| Superpowers superpowers | A1 Skills-only | Seed | 207,067 | 3 | claude-code | | Enforces spec-first, TDD, and subagent-reviewed development as mandatory automatic workflows rather than optional practices. |
| Spec Kit spec-kit | A2 Mirror cmd+skill | Seed | 105,966 | 1.5 | copilot | MCP | Turns a natural-language feature description into a complete, versioned, AI-executable specification pipeline installable for 30+ agents in one CLI command. |
| Claude-Flow / Ruflo claude-flow | A6 Multi-agent orchestrator | Seed | 55,218 | 3.6 | claude-code | MCP | Eliminates single-agent context limits and sequential bottlenecks by orchestrating fault-tolerant swarms of specialized AI agents with shared persistent memory. |
| OpenSpec openspec | A2 Mirror cmd+skill | Seed | 50,822 | 1.5 | unknown | | Adds a lightweight spec layer so AI coding assistants and humans agree on what to build before any code is written. |
| BMAD-METHOD bmad-method | A4 Markdown scaffold | Seed | 48,044 | 2 | claude-code | | Provides a full agile delivery lifecycle with named expert-persona AI collaborators that elicit the human's best thinking rather than substituting AI opinion for it. |
| Taskmaster AI taskmaster-ai | A3 MCP-anchored | Seed | 27,254 | 2 | cursor | MCP | Converts a PRD into a dependency-ordered JSON task graph that AI coding agents execute one task at a time, eliminating context overload in long-running AI-driven development sessions. |
| Agent OS agent-os | A4 Markdown scaffold | Seed | 4,646 | 1 | claude-code | | Extracts implicit codebase conventions into token-efficient markdown standards files and injects them selectively into AI agent context windows so agents build the way your team actually builds. |
| Claude Conductor claude-conductor | A4 Markdown scaffold | Seed | 367 | 1.5 | claude-code | | Gives Claude Code a persistent, cross-linked, auto-analyzed documentation system so it retains codebase context across sessions. |
| Spec-Driver (Greenfield Spec-Driven Development) spec-driver | A4 Markdown scaffold | Seed | 25 | 2 | claude-code | | Prevents spec rot in AI-assisted development by making implementation changes flow back into evergreen, authoritative specs via audits, drift ledgers, and design revisions. |
| ccmemory ccmemory | A3 MCP-anchored | Seed | 1 | 1.5 | claude-code | MCP | Accumulates decisions, corrections, and failed approaches from Claude Code sessions into a queryable Neo4j graph so each new session starts with full project context rather than from zero. |
| DeerFlow deerflow | A6 Multi-agent orchestrator | Tier A | 69,608 | 6 | web-ui | UI | Long-horizon superagent that researches, codes, and creates by orchestrating parallel sub-agents with isolated contexts in Docker sandboxes, driven by progressive SKILL.md skill loading. |
| oh-my-openagent (omo) oh-my-openagent | A6 Multi-agent orchestrator | Tier A | 59,562 | 8 | opencode | CLI UI MCP | Multi-provider AI agent orchestration for OpenCode: escape vendor lock-in by routing Sisyphus (Claude/Kimi/GLM) and Hephaestus (GPT-5.5) as a coordinated team. |
| oh-my-codex (Yeachan-Heo) oh-my-codex-yeachan | A6 Multi-agent orchestrator | Tier A | 29,662 | 8 | codex-cli | CLI UI MCP | Add structured planning, durable state, and multi-agent team coordination to OpenAI Codex CLI without replacing it. |
| Symphony (OpenAI) symphony-openai | A7 PR-lifecycle platform | Tier A | 24,645 | 4.5 | codex | CLI UI | A language-agnostic specification and Elixir reference daemon that continuously polls Linear and dispatches isolated Codex sessions per issue, letting teams manage work instead of supervising coding agents. |
| Codex Plugin for Claude Code (openai) codex-plugin-cc | A6 Multi-agent orchestrator | Tier A | 19,674 | 5.2 | claude-code | | Use OpenAI Codex as a reviewer and task executor from inside an active Claude Code session without leaving the CC workflow. |
| Compound Engineering compound-engineering | A8 Cross-runtime harness | Tier A | 17,199 | 8.1 | claude-code | CLI | Make each unit of engineering work compound into easier future work via brainstorm→plan→execute→review→learn cycles. |
| Context Mode context-mode | A19 Token-efficient encoding | Tier A | 15,710 | 3 | claude-code | CLI UI MCP | Keeps raw tool output data out of the context window via sandbox execution and SQLite+FTS5 session indexing, reducing context consumption by up to 98% while providing selective BM25-based session continuity after compaction. |
| Gas Town gas-town | A6 Multi-agent orchestrator | Tier A | 15,581 | 7.4 | claude-code | CLI UI | Coordinate 20-30 concurrent AI coding agents on shared repos with persistent work state, crash recovery, and autonomous health monitoring. |
| Open Multi-Agent open-multi-agent | A10 Memory engine | Tier A | 6,251 | 4 | node-js-backends | CLI UI | Give a natural-language goal to a coordinator agent and get a dynamically decomposed, parallelized task DAG executed by specialized worker agents — zero graph wiring required. |
| Pimzino spec-workflow-mcp spec-workflow-mcp | A3 MCP-anchored | Tier A | 4,199 | 3.6 | claude-code | CLI UI MCP | MCP server providing spec-driven development workflow with dashboard-backed approval gates, implementation logging, and VSCode extension for multi-tool AI clients. |
| claude-code-spec-workflow (Pimzino) claude-code-spec-workflow | A6 Multi-agent orchestrator | Tier A | 3,744 | 2.4 | claude-code | CLI UI | Two-workflow Claude Code spec harness: structured feature development plus dedicated bug-fix pipeline, with a real-time dashboard and Tauri desktop app for team visibility. |
| Harness (revfactory) — Team-Architecture Factory revfactory-harness | A15 Harness generator | Tier A | 3,552 | 2 | claude-code | | Generates domain-specific agent teams and their skills from a natural language domain description, using one of 6 pre-defined architecture patterns. |
| cc-sdd (gotalab) cc-sdd | A8 Cross-runtime harness | Tier A | 3,394 | 4.2 | claude-code | CLI | Portable Kiro-style SDD harness for 8 AI agents: discovery-routed spec creation with boundary-annotated tasks and autonomous subagent-per-task TDD implementation. |
| Amplifier (microsoft) amplifier | A8 Cross-runtime harness | Tier A | 3,095 | 1 | anthropic-api | CLI | Ultra-thin Python kernel (~2,600 lines) with formal module protocol contracts and a Git-based bundle marketplace, providing a Linux-kernel-style extensible AI assistant platform. |
| CodeMachine CLI codemachine-cli | A16 Cross-vendor router | Tier A | 2,488 | 4.5 | claude-code | CLI UI MCP | JavaScript-DSL workflow orchestration engine that captures repeatable AI coding agent workflows with tracks, condition groups, and long-running persistence across Claude Code, Codex, Cursor, and Gemini. |
| APM (Agentic Project Management) apm-agentic-project-mgmt | A8 Cross-runtime harness | Tier A | 2,296 | 6.5 | claude-code | CLI | Manages complex multi-session software projects by coordinating specialized AI agents (Planner/Manager/Workers) with human-mediated message relay and structured Handoff when context fills. |
| Chachamaru Claude Code Harness chachamaru-cc-harness | A17 Compiled enforcement | Tier A | 1,611 | 9.5 | claude-code | CLI | Turns raw Claude Code sessions into a disciplined Plan→Work→Review→Release delivery loop with spec contracts and worktree-isolated agent teams. |
| Agentic Coding Flywheel Setup (ACFS) flywheel-sdd | A11 Governance | Tier A | 1,488 | 3.2 | claude-code | CLI UI | Take a complete beginner from laptop to three AI coding agents running on a VPS in 30 minutes via an idempotent manifest-driven Bash installer. |
| Haft haft | A8 Cross-runtime harness | Tier A | 1,333 | 4.4 | claude-code | CLI UI MCP | Engineering reasoning governor that enforces First Principles Framing — frame, compare under parity, decide with falsifiable contracts, detect stale evidence — before AI agents execute. |
| quint-code (Haft) quint-code | A8 Cross-runtime harness | Tier A | 1,333 | 6.7 | claude-code | CLI UI MCP | Governs AI agent execution through formal decision contracts, parity-enforced comparisons, evidence-decay scoring, and bounded WorkCommissions — ensuring agents operate within explicit invariants rather than ad-hoc prompts. |
| Codex Integration for Claude Code (skill-codex) codex-integration-skill | A4 Markdown scaffold | Tier A | 1,269 | 3 | claude-code | | Single Claude Code skill that handles Codex CLI invocation correctly (stdin blocking, thinking token suppression, session resume) with a peer-AI critical evaluation protocol. |
| Spec Kitty spec-kitty | A8 Cross-runtime harness | Tier A | 1,263 | 5 | claude-code | CLI UI | Delivers a complete spec-driven agentic workflow across 17 AI agents with worktree isolation, lane-tracked parallel execution, and retrospective-based workflow learning. |
| Claude Command Suite (qdhenry) claude-command-suite-qdhenry | A6 Multi-agent orchestrator | Tier A | 1,258 | 3.1 | claude-code | | 216+ slash commands across 26 domain namespaces + 12 skills + 54 agents covering full-stack development, DevOps, security, media processing, browser AI, and business scenario modeling. |
| agentflow (berabuddies) agentflow-berabuddies | A8 Cross-runtime harness | Tier A | 1,255 | 2.5 | codex | CLI | Python DSL for building multi-agent DAG pipelines with >> operator, Jinja2 prompt chaining, fanout/merge parallelism, iterative LLM-as-judge cycles, and native EC2/ECS/SSH remote execution. |
| Agor agor | A10 Memory engine | Tier A | 1,221 | 5.5 | claude-code | CLI UI MCP | Team-first spatial canvas for AI agent work — branches as cards on a Figma-style board with zone-triggered prompts, real-time multiplayer, and session genealogy. |
| RooFlow rooflow | A10 Memory engine | Tier A | 1,159 | 4 | roo-code | MCP | YAML-based system prompt replacement for Roo Code's Footgun feature with a persistent file-based Memory Bank for cross-session context continuity. |
| cursorrules v5 (kinopeee) cursorrules-kinopeee | A2 Mirror cmd+skill | Tier A | 1,118 | 1.4 | cursor | | Bilingual (ja/en) Cursor rule set with tricolor task classification, security-first prompt injection defense, and structured git workflow commands. |
| Bhartendu-Kumar/rules_template rules-template-stackable | A8 Cross-runtime harness | Tier A | 1,059 | 3 | cursor | MCP | Cross-platform rule template for Cursor + Cline + Roo Code using symbolic links for a shared single source of truth with Agile-inspired SDLC workflow. |
| thClaws thclaws | A8 Cross-runtime harness | Tier A | 1,043 | 7 | thclaws | CLI UI | One native Rust binary that is a complete sovereign AI agent workspace — GUI, CLI, webapp, scheduler, knowledge base, and multi-agent orchestration — without cloud dependencies. |
| MoAI-ADK moai-adk | A18 Self-evolving | Tier A | 1,030 | 8.1 | claude-code | CLI MCP | Implements Harness Engineering as a Go-binary-installed Claude Code environment with auto-TDD/DDD methodology selection, 20-event hook coverage, self-evolving V3R4 harness, and MX-tag in-code context annotations. |
| oh-my-agent oh-my-agent | A8 Cross-runtime harness | Tier A | 1,020 | 5.8 | Claude Code | CLI UI MCP | Portable multi-agent harness that models an engineering team as specialized agents and projects them across 27+ AI tools from a single .agents/ source of truth |
| Superpowers Marketplace superpowers-marketplace | A99 Unclassified | Tier A | 1,006 | 0 | claude-code | | Single-endpoint Claude Code plugin marketplace for the superpowers plugin ecosystem. |
| GET SHIT DONE (GSD) get-shit-done | A8 Cross-runtime harness | Tier A | 943 | 4.5 | claude-code | CLI | Prevent AI context rot by keeping the orchestrator lean (~15% context) and running all implementation in parallel fresh 200k-token subagent contexts. |
| Chorus chorus | A6 Multi-agent orchestrator | Tier A | 915 | 3.6 | Claude Code | CLI UI MCP | Full-stack agent harness with Reversed Conversation philosophy — AI proposes, humans verify — enforced via a permission-gated MCP server and production web UI |
| Cursor Rules Pack v2 cursor-rules-pack-v2 | A6 Multi-agent orchestrator | Tier A | 853 | 8.6 | cursor | CLI | Official Cursor plugin marketplace providing CI workflows, engineering principles, parallel cloud-agent orchestration, transcript-based memory, and iterative completion loops. |
| your-claude-engineer (coleam00) coleam00-your-claude-engineer | A6 Multi-agent orchestrator | Tier A | 850 | 4.9 | terminal | MCP | Autonomous software factory that takes a plain-text app spec and delivers a running web app with Linear issues tracked, GitHub PRs created, and Slack notifications sent — enforcing a screenshot evidence gate before any feature is marked complete. |
| AgentSys (avifenesh/awesome-slash) awesome-slash-commands | A8 Cross-runtime harness | Tier A | 823 | 4.1 | claude-code | CLI | Orchestrate everything around code-writing — task selection, branch management, review, PR, merge, memory — via structured pipelines with gated phases and persistent state. |
| codex-autorunner (CAR) codex-autorunner | A10 Memory engine | Tier A | 809 | 6 | openai-codex | CLI UI | Ticket-driven background daemon that orchestrates Codex (or any ACP agent) against a work queue, notifying humans via Telegram/Discord only when agents are stuck. |
| Roo Commander (jezweb) roo-commander | A6 Multi-agent orchestrator | Tier A | 725 | 4 | roo-code | CLI | Bridge Claude Code's 60+ production-tested skills to Roo Code via an orchestrator mode and CLI skill-discovery tool. |
| cli-agent-orchestrator (awslabs) cli-agent-orchestrator | A7 PR-lifecycle platform | Tier A | 634 | 3 | kiro-cli | CLI UI MCP | tmux-based supervisor-worker orchestration across 7 AI coding CLIs, with MCP coordination primitives, persistent memory, scheduled flows, and a bundled React dashboard — without replacing any existing CLI tool. |
| flow-next flow-next | A8 Cross-runtime harness | Tier A | 615 | 7.9 | claude-code | CLI UI | Enforce spec-first task hygiene with fresh-context workers, source-tagged capture, and cross-model review gates to prevent context bleed and hallucinated requirements. |
| Nimbalyst nimbalyst | A12 UI passthrough | Tier A | 604 | 4.3 | codex-cli | UI | Replace terminal-switching and raw diff review with a WYSIWYG visual editor and session manager where humans and AI coding agents collaborate on the same documents. |
| cline-memory-bank (nickbaumann98) cline-memory-bank-nickbaumann | A13 Methodology | Tier A | 581 | 0.5 | cline | | Custom instructions + 6-file hierarchical Markdown memory bank so Cline maintains full project context across sessions, with a .clinerules learning journal for behavioral patterns. |
| adversarial-spec adversarial-spec | A6 Multi-agent orchestrator | Tier A | 546 | 1.9 | claude-code | | Iteratively refines product specifications through multi-LLM adversarial debate until all models reach consensus, surfacing gaps and edge cases that any single model review would miss. |
| SwarmClaw swarmclaw | A10 Memory engine | Tier A | 523 | 5.5 | claude-code | CLI UI MCP | Self-hosted autonomous agent runtime with persistent memory, org-chart delegation, dreaming memory consolidation, and 23+ LLM providers — a practical Claude Code and LangChain alternative. |
| MateClaw mateclaw | A8 Cross-runtime harness | Tier A | 512 | 5.5 | web-console | UI | Enterprise multi-agent server platform: always-on, multi-vendor failover, RBAC, approval gates, audit trail, and 8 IM channel adapters in a single Spring Boot JAR. |
| Bernstein bernstein | A3 MCP-anchored | Tier A | 460 | 8.5 | claude-code | CLI UI MCP | Govern parallel CLI coding agents with a deterministic Python scheduler, HMAC-chained audit trail, and compliance-ready signed lineage so multi-agent runs are reproducible and auditability is provable. |
| Bernstein (sipyourdrink-ltd) bernstein-sipyourdrink | A4 Markdown scaffold | Tier A | 460 | 0 | unknown | | Alias entry — same repo as bernstein slug. See spec-system/bernstein/ for full analysis. |
| mcp-server-spec-driven-development mcp-server-spec-driven-development | A11 Governance | Tier A | 430 | 0.5 | vscode-copilot | MCP | Provides a minimal MCP server with three EARS-format prompts that enforce the requirements→design→code spec-driven development pipeline in any MCP-compatible AI tool. |
| Claudiomiro claudiomiro | A10 Memory engine | Tier A | 412 | 5.5 | claude-code | CLI | Fully autonomous parallel coding agent that decomposes prompts via DAG, executes tasks simultaneously with multi-provider support, learns from reflections, and commits tested production-ready code. |
| SpecPulse specpulse | A8 Cross-runtime harness | Tier A | 385 | 0.5 | claude-code | CLI | Deploys an identical spec-driven workflow across 8 AI platforms using a CLI-first scaffold that the AI then populates. |
| FPF (First Principles Framework) fpf | A13 Methodology | Tier A | 372 | 3.6 | any | | Provides a formal pattern language for making reasoning explicit, traceable, and publishable in mixed human/AI engineering work — addressing semantic drift, premature comparison, temporal claim staleness, and causal misuse. |
| windsurfrules v5 (kinopeee) windsurfrules-kinopeee | A2 Mirror cmd+skill | Tier A | 364 | 1.4 | windsurf | | Windsurf/Antigravity port of cursorrules v5 — same tricolor task classification and injection defense, translated to Windsurf's .windsurf/rules/ format and tool naming. |
| OpenClaw.NET openclaw-net | A11 Governance | Tier A | 347 | 7 | web-dashboard | CLI UI MCP | Self-hosted NativeAOT .NET agent gateway with inspectable Passive Harness Contracts, Evidence Bundles, and a Governance Ledger for observable, auditable agent work without changing default behavior. |
| Superpowers Lab superpowers-lab | A1 Skills-only | Tier A | 344 | 1 | claude-code | | Experimental skill incubator extending Superpowers with interactive CLI automation, on-demand MCP discovery, semantic deduplication, and infrastructure management. |
| Smart Ralph smart-ralph | A6 Multi-agent orchestrator | Tier A | 339 | 3.5 | claude-code | | Turn a vague feature idea into structured specs and then execute them task-by-task with fresh subagent context, without losing state across sessions. |
| metaswarm metaswarm | A10 Memory engine | Tier A | 284 | 9.1 | claude-code | CLI | Production-tested multi-agent orchestration with self-improving knowledge base, recursive Swarm Coordinator architecture, and mandatory cross-model adversarial review — supporting Claude Code, Gemini CLI, and Codex CLI simultaneously. |
| Conductor (MadAppGang) claude-conductor-madappgang | A4 Markdown scaffold | Tier A | 280 | 5.5 | claude-code | | CDD plugin inside a 14-plugin professional consulting marketplace; uniquely uses Claude Code's native Tasks API and git notes for plan tracking and audit trail. |
| DashClaw dashclaw | A11 Governance | Tier A | 268 | 5 | claude-code | CLI UI MCP | Full-stack AI agent governance platform: pre-action policy enforcement, multi-channel human approval routing, durable finality, audit trails, and Code Sessions optimizer. |
| Symphony (sincover) symphony-sincover | A6 Multi-agent orchestrator | Tier A | 260 | 3.6 | roo-code | | 12-role musical-metaphor hierarchy (Composer→Score→Conductor→Performer) for Roo Code, with three automation levels, Mermaid visualization, and append-only communication logs. |
| LeanSpec lean-spec | A3 MCP-anchored | Tier A | 252 | 1 | claude-code | CLI UI MCP | Provides a unified spec CLI and MCP server over any existing spec backend (markdown, GitHub Issues, ADO), making spec-driven development tool-agnostic without requiring teams to change their workflow. |
| am-will/swarms am-will-swarms | A10 Memory engine | Tier A | 203 | 4.1 | claude-code | | Context-persistent parallel multi-agent execution: the orchestrator that planned the work stays alive across all execution waves, pre-loading subagents with exactly the context they need. |
| ralphy-openspec ralphy-openspec | A11 Governance | Tier A | 186 | 6.9 | claude-code | CLI | Combines OpenSpec spec governance with Ralph Loop iterative execution and a SQLite run ledger to provide crash-recoverable, budget-tracked AI coding sessions. |
| spec-kit-command-cursor (madebyaris) spec-kit-command-cursor | A6 Multi-agent orchestrator | Tier A | 182 | 5.9 | cursor | | Cursor-native SDD toolkit with async background subagent trees, parallel DAG execution with file-conflict detection, checkpoint-resume, and downstream spec propagation. |
| Custom-Modes-Roo-Code (jtgsystems) custom-modes-roo-code | A4 Markdown scaffold | Tier A | 173 | 1.6 | roo-code | | 225-agent configuration library for Roo Code covering every major software development domain with 2026 security-first standards. |
| Claude Swarm (affaan-m) claude-swarm | A6 Multi-agent orchestrator | Tier A | 172 | 4.1 | claude-code | CLI UI | Decompose complex tasks via Opus into a dependency DAG, execute waves of Haiku workers in parallel, then Opus quality-gates the combined output — mirroring real engineering team structure. |
| Spec-Driven Agentic Development spec-driven-agentic-development | A6 Multi-agent orchestrator | Tier A | 172 | 1 | claude-code | | Generates EARS-formatted specs via parallel codebase exploration agents before any implementation begins, ensuring specs reflect actual code patterns. |
| OpenCode Orchestrator opencode-orchestrator | A6 Multi-agent orchestrator | Tier A | 167 | 4.2 | opencode | | Hub-and-spoke hierarchical orchestration for OpenCode with Chase-Lev work-stealing, MVCC state sync, and evidence-based anti-hallucination mandates. |
| sd0x-dev-flow sd0x-dev-flow | A8 Cross-runtime harness | Tier A | 157 | 6.4 | claude-code | | Reference implementation of harness engineering for Claude Code — hook-enforced dual review, sentinel-driven state machine, and fail-closed safety where the AI cannot skip quality gates. |
| specs.md (AI-DLC) ai-dlc-specs-md | A8 Cross-runtime harness | Tier A | 156 | 3.5 | claude-code | CLI UI | Ships three selectable SDD methodologies (Simple/FIRE/AI-DLC) in one npm package so teams can graduate from lightweight to full lifecycle orchestration without switching toolchains. |
| Conductor (microsoft) conductor-microsoft | A8 Cross-runtime harness | Tier A | 156 | 3.7 | conductor-cli | CLI UI | Deterministic YAML-defined multi-agent workflow engine with Jinja2 routing, parallel execution, multi-model support, and a built-in real-time web dashboard. |
| GAAI Framework gaai-framework | A11 Governance | Tier A | 142 | 5 | claude-code | UI | Dual-track governed delivery system: Discovery defines scope, Delivery executes in isolated OS processes, backlog-as-contract prevents scope drift, persistent memory enables cross-session learning. |
| Superpowers: Developing for Claude Code superpowers-developing-for-claude-code | A4 Markdown scaffold | Tier A | 129 | 0.5 | claude-code | | Gives Claude Code comprehensive, self-updating knowledge of the Claude Code platform for building plugins, skills, hooks, commands, and MCP servers. |
| Specs Workflow MCP specs-workflow-mcp | A3 MCP-anchored | Tier A | 127 | 1.8 | claude-code | MCP | Enforces Requirements → Design → Tasks workflow via a single MCP tool with persistent JSON progress tracking that survives session restarts. |
| hankweave hankweave | A8 Cross-runtime harness | Tier A | 123 | 5.5 | Claude Code | CLI UI | Production operations runtime for executing frozen, long-horizon agentic programs (hanks) reliably, with single-threaded execution, git checkpointing, and sentinel monitoring |
| HexAgent hexagent | A8 Cross-runtime harness | Tier A | 122 | 3 | python-library | | Gives any LLM a computer via a runtime-computer isolation protocol — the harness never shares its keys or config with the agent. |
| Utah (Inngest) utah-inngest | A4 Markdown scaffold | Tier A | 116 | 5 | slack | | Durable Inngest-powered personal agent where every LLM call and tool execution is a checkpointed step with automatic retries, singleton concurrency, and cancel-on-new-message — no ngrok needed. |
| Code Conductor (ryanmac) code-conductor | A6 Multi-agent orchestrator | Tier A | 101 | 5.4 | claude-code | CLI | Decentralized swarm orchestration for parallel Claude Code agents using GitHub Issues as the atomic task queue and git worktrees for zero-conflict isolation. |
| revfactory Claude Code Harness (A/B Research) revfactory-cc-harness | A6 Multi-agent orchestrator | Tier A | 98 | 3.7 | claude-code | | Empirically proves that Claude Code pre-configuration improves output quality by 60% across 15 software engineering tasks. |
| AI Governor Framework (Fr-e-d) ai-governor-framework | A11 Governance | Tier A | 78 | 1.4 | cursor | | Provides structured governance protocols and in-repo rules to transform AI assistants into disciplined engineering partners that respect project architecture and coding standards. |
| Spartan AI Toolkit spartan-ai-toolkit | A11 Governance | Tier A | 72 | 5.6 | claude-code | CLI | Engineering discipline layer delivering 73 commands, 9 reviewer agents, 8 stack profiles, and mandatory git worktree + quality gate enforcement for every AI-generated feature. |
| fspec (SENGAC) fspec | A4 Markdown scaffold | Tier A | 69 | 7.5 | claude-code | CLI UI MCP | Provides complete ACDD infrastructure for AI agents: Gherkin spec management, Example Mapping facilitation, TDD state machine enforcement, Git checkpoints for safe recovery, and coverage tracking linking scenarios to tests to implementation. |
| oh-my-codex (scalarian) oh-my-codex-scalarian | A6 Multi-agent orchestrator | Tier A | 65 | 6 | codex-cli | CLI UI MCP | A focused Codex-only workflow runtime with tmux team execution, MCP servers, and durable .omx/ state — the v2 vision without Claude bridge. |
| orchestr8 orchestr8 | A14 JIT loader | Tier A | 65 | 0 | claude-code | MCP | JIT knowledge loading plugin for Claude Code — 383 resource fragments (154 domain agents, 103 skills, 36 workflows, 32 patterns) loaded on-demand via MCP with 95-98% token reduction vs upfront loading; AITMPL community registry provides 400+ shared resources. |
| IBM IaC Spec Kit iac-spec-kit | A11 Governance | Tier A | 64 | 1.5 | claude-code | CLI | Adapts SDD methodology to IaC by separating generic infrastructure specifications from cloud-specific implementation plans, enabling multi-cloud comparison and team readability. |
| barkain/claude-code-workflow-orchestration barkain-workflow-orch | A6 Multi-agent orchestrator | Tier A | 60 | 3.9 | claude-code | | Enforce Director/Coder split via adaptive hook-based delegation nudges, native plan mode as the handoff boundary, and 8 specialized agents for domain execution. |
| cline-mcp-memory-bank (dazeb) cline-mcp-memory-bank | A10 Memory engine | Tier A | 60 | 0.5 | cline | MCP | Provides four structured Markdown files (projectContext, activeContext, progress, decisionLog) managed via an MCP server so Cline retains project context between sessions. |
| CLAUDER clauder | A11 Governance | Tier A | 58 | 5.6 | claude-code | CLI UI MCP | Safety-first Claude Code toolkit with 14 PreToolUse/PostToolUse hooks, SQLite audit trail, web tracer, and 67 domain expansion agents. |
| Claude Conductor (rbarcante) claude-conductor-rbarcante | A6 Multi-agent orchestrator | Tier A | 53 | 6 | claude-code | | Context-Driven Development plugin that enforces a spec-plan-implement lifecycle with Python-CLI context management, pattern library, and ADR logging for Claude Code. |
| aistack aistack | A6 Multi-agent orchestrator | Tier A | 51 | 6.5 | claude-code | CLI UI MCP | Coordinates 11 specialized AI agents with adversarial code review, JWT agent identity, and 6-provider multi-model routing against shared SQLite memory. |
| MetaSpec (ACNet-AI) metaspec-acnet | A8 Cross-runtime harness | Tier A | 47 | 1.4 | claude-code | CLI | Generates complete spec-driven toolkits for any domain from a single command, enabling the creation of domain-specific specification systems rather than applying generic SDD to every project. |
| shinpr/agentic-code shinpr-agentic-code | A8 Cross-runtime harness | Tier A | 46 | 2 | cursor | CLI | AGENTS.md-based framework that enforces TDD, Plan Injection gates, and progressive skill loading across Cursor, Codex CLI, and Gemini CLI. |
| codex-spec (shenli) codex-spec | A99 Unclassified | Tier A | 45 | 0.5 | codex-cli | CLI | Automates the spec-to-code pipeline for OpenAI Codex by generating specifications, requirements, plans, and tasks from natural language feature descriptions. |
| cc-plugin-codex (sendbird) cc-plugin-codex-sendbird | A6 Multi-agent orchestrator | Tier A | 42 | 5.2 | openai-codex | | Use Claude Code for review and rescue from inside a Codex session — the semantic inverse of codex-plugin-cc, with Codex as host and Claude as the worker. |
| REAP (c-d-cc/reap) reap | A18 Self-evolving | Tier A | 41 | 7 | claude-code | CLI | Prevent context loss, scattered development, and forgotten lessons through a generation-based lifecycle where AI and human co-evolve a typed knowledge base (Genome + Environment) across successive iterations. |
| Tessl tessl | A8 Cross-runtime harness | Tier A | 38 | 1.9 | claude-code | CLI | Steering tile that enforces spec-before-code methodology via versioned skills, always-on rules, and an evaluation harness with 9 graded scenarios. |
| Tessl SDD Tile tessl-sdd-tile | A8 Cross-runtime harness | Tier A | 38 | 1.9 | claude-code | | Provides a skills+rules+evals tile for spec-driven development with one-question-at-a-time requirement gathering, explicit stakeholder approval gates, and inline [@test] link traceability in .spec.md files. |
| Uniswap AI Toolkit ai-toolkit-uniswap | A4 Markdown scaffold | Tier A | 37 | 2 | Claude Code | | Organizational Claude Code plugin marketplace that packages Uniswap Labs' engineering culture (feature planning, PR workflows, codebase navigation, Slack/Notion knowledge management) into reusable AI agents, skills, and hooks |
| VNX Orchestration vnx-orchestration | A11 Governance | Tier A | 34 | 7.5 | claude-code | CLI UI | Governance-first multi-agent orchestration: every agent action produces a receipt, quality gates use deterministic file-based verdicts (not LLM judgment), and full provenance is maintained across all providers. |
| GreatCTO greatcto | A11 Governance | Tier A | 32 | 7.1 | claude-code | CLI UI MCP | Full SDLC pipeline (57 specialist agents, 2 human gates, 12 jurisdiction detectors, 33+ compliance frameworks) for solo founders — architect/critics/reviewers/QA/security/devops run automatically; human decides at plan and ship. |
| SuperBeads Universal Framework (Wiggum Flavour) superbeads-wiggum | A4 Markdown scaffold | Tier A | 31 | 2.4 | claude-code | CLI UI | Bring Beads task tracking, atomic git commits, session memory, and domain-specific verification to Claude Code via a universal CLI-installed scaffold. |
| Forge (LucasDuys) forge-lucasduys | A11 Governance | Tier A | 29 | 7.4 | claude-code | CLI | Replace the developer as the state machine by running tasks in isolated git worktrees with TDD, automatic backpropagation of runtime failures into specs, and a stop-hook loop that drives everything. |
| GoopSpec goopspec | A6 Multi-agent orchestrator | Tier A | 29 | 8.6 | opencode | CLI UI MCP | Enforces spec-driven development in OpenCode via a 5-phase workflow with two mandatory human contract gates, a 16-agent multi-model orchestration system, and a memory pipeline that learns from every completed milestone. |
| cc-spec-driven cc-spec-driven | A21 LLM-evaluated hooks | Tier A | 27 | 4.9 | claude-code | | Tracks specification changes as CRs through a draft→confirm→RC→archive lifecycle with LLM-evaluated hooks that enforce checklist completion and bidirectional dependency graph integrity. |
| FlowCoder flowcoder | A4 Markdown scaffold | Tier A | 26 | 1.9 | claude-code | CLI UI | Build complex multi-step autonomous agent workflows visually via a flowchart builder — no orchestration code required. |
| WORCA worca | A11 Governance | Tier A | 26 | 9.5 | claude-code | CLI UI | 9-stage autonomous development pipeline with OS-level governance enforcement, per-agent model routing, real-time WebSocket dashboard, and fleet/workspace orchestration for parallel multi-project runs. |
| Conductor (fcoury) conductor-fcoury | A4 Markdown scaffold | Tier A | 25 | 4.5 | claude-code | | Cross-tool CDD implementation running identical methodology on Claude Code, Gemini CLI, and Codex CLI with an auto-activating ambient skill. |
| openspec-mcp (Lumiaqian) openspec-mcp | A3 MCP-anchored | Tier A | 24 | 2.6 | claude-code | CLI UI MCP | Exposes the full OpenSpec lifecycle (changes, approvals, reviews, tasks, validation) as MCP tools plus a real-time web dashboard. |
| SWORDSwarm sword-swarm | A8 Cross-runtime harness | Tier A | 24 | 1.8 | claude-code | | 88-agent corporate-hierarchy orchestration system with Intel NPU hardware acceleration and multi-IDE support for enterprise-grade parallel AI task execution. |
| roomodes (sellisd) roomodes-sellisd | A4 Markdown scaffold | Tier A | 23 | 0 | roo-code | MCP | Simulate a full agile scrum team (PO, SM, Dev, QA) in Roo Code using per-role Jira-integrated personas. |
| HOTL Plugin hotl-plugin | A11 Governance | Tier A | 22 | 6 | claude-code | | Human-on-the-loop structured workflows that enforce design-before-code, resumable execution with state persistence, and explicit branch disposition — across 5 AI coding tools. |
| Pluqqy pluqqy | A8 Cross-runtime harness | Tier A | 21 | 0.5 | claude-code | CLI UI | Compose reusable context/prompt/rule components into named pipelines that activate with one command, enabling instant context switching between development tasks. |
| do-it do-it | A6 Multi-agent orchestrator | Tier A | 20 | 6.6 | codex-cli | CLI | Stop asking AI agents to remember process — install risk-routing, evidence-gated completion, and explicit subagent contracts as automatic hooks. |
| openspec-for-copilot openspec-for-copilot | A5 Closed IDE | Tier A | 18 | 0.5 | github-copilot | UI | Brings OpenSpec spec-driven development to GitHub Copilot Chat users via a VS Code extension with sidebar spec management, Handlebars-compiled prompts, and CodeLens task runner. |
| openspec-ui (ToruAI) openspec-ui | A12 UI passthrough | Tier A | 18 | 0.5 | none | CLI UI | Multi-repository dashboard that reads OpenSpec directories and displays them as a real-time Kanban board — mission control for AI-assisted development with no AI interaction itself. |
| AI Engineering Harness adrielp-ai-engineering-harness | A8 Cross-runtime harness | Tier A | 16 | 2.5 | claude-code | CLI | Translates a shared set of context-engineering patterns into native formats for 4 different AI coding tools via a Deno CLI installer. |
| ching-kuo/claude-codex ching-kuo-claude-codex | A4 Markdown scaffold | Tier A | 16 | 4.5 | claude-code | MCP | Bidirectional Claude↔Codex critique loop via MCP where Claude plans/implements and Codex audits/reviews with APPROVED/WARNING/BLOCKED verdicts. |
| cline-spec-kit-workflows (JRedeker) cline-spec-kit-workflows | A13 Methodology | Tier A | 16 | 4.4 | cline | | Ports GitHub Spec Kit's spec-driven development workflow to Cline via eight procedural workflow files. |
| Vibook (pinqy520) roo-code-vibook | A4 Markdown scaffold | Tier A | 16 | 1 | roo-code | | Generate an AI-optimized documentation knowledge base for any existing codebase to enable efficient AI-assisted development and maintenance. |
| CCCT (IgorWarzocha) ccct | A4 Markdown scaffold | Tier A | 15 | 2.5 | claude-code | | Professional development workflow chain (setup→explore→PRD→tasks→dev→review) with complexity-driven task breakdown and dual-tracking for Claude Code. |
| Claude Conductor (lackeyjb) claude-conductor-lackeyjb | A6 Multi-agent orchestrator | Tier A | 14 | 5.9 | claude-code | | Simplified Context-Driven Development plugin with 3-agent specialization (planner/implementer/reviewer) and lifecycle hooks for passive context management. |
| oh-my-codex (sigridjineth) oh-my-codex-sigridjineth | A6 Multi-agent orchestrator | Tier A | 14 | 5.8 | codex-cli | CLI | A personal Codex plugin extension adding tier-based model routing, LSP/AST tooling, and notification integrations to the canonical oh-my-codex workflow. |
| OpenSpec-Zed openspec-zed | A5 Closed IDE | Tier A | 14 | 0 | zed | UI | Adds OpenSpec slash commands to the Zed editor's AI Assistant panel so Zed users can initiate spec-driven workflows without leaving the editor. |
| SpecPact specpact | A10 Memory engine | Tier A | 13 | 2.2 | claude-code | CLI | Provide permanent, tiered spec contracts with a Memory Bank so AI sessions always have project context and always implement against explicit contracts rather than inferred intent. |
| Cursor Agentic Development Pipeline cursor-agentic-pipeline | A4 Markdown scaffold | Tier A | 12 | 1 | github-copilot | MCP | Automate building an Obsidian knowledge base of Laravel packages, skills, and MCP servers using GitHub Copilot cloud agents. |
| shinpr/codex-workflows shinpr-codex-workflows | A6 Multi-agent orchestrator | Tier A | 12 | 7.3 | openai-codex | CLI | Codex-native 13-stage pipeline with context-isolated subagents that prevents assumption drift in large multi-file changes. |
| Aigon aigon | A8 Cross-runtime harness | Tier A | 10 | 7.4 | claude-code | CLI UI | Orchestrates 7 different AI coding agent CLIs from a single Kanban workflow with per-feature worktree isolation, Fleet mode for parallel agent comparison, and a local web dashboard. |
| OneBrain onebrain | A8 Cross-runtime harness | Tier A | 10 | 3 | claude-code | CLI UI | Give AI agents persistent memory, 34 skills, and personal calibration via an Obsidian vault so every session picks up exactly where the last one left off — across any harness. |
| Cline ACP cline-acp | A8 Cross-runtime harness | Tier A | 9 | 1 | zed | CLI | ACP protocol adapter that exposes Cline's coding capabilities to non-VS-Code editors like Zed. |
| Agent Harness Skill (Harness Engineering) ldzhouquan-agent-harness-skill | A4 Markdown scaffold | Tier A | 9 | 4.2 | claude-code | | Systematic Harness Engineering methodology with Iron Laws, Reflexion Loop, and 5-second session-amnesia recovery protocol. |
| SpecD specd-sdd | A8 Cross-runtime harness | Tier A | 9 | 4.4 | claude-code | CLI MCP | Pre-assembles deterministic context packages for agents at each lifecycle step, combined with a multi-language code graph and customizable schema that defines artifact workflow per project. |
| SuperSpec (danielhanold) superspec | A6 Multi-agent orchestrator | Tier A | 8 | 8.5 | claude-code | | Eliminates the duplicate artifacts and manual skill selection overhead of using OpenSpec and Superpowers together by providing a schema that enforces their correct integration sequence. |
| Codex Harness MCP chapzin-codex-harness-mcp | A18 Self-evolving | Tier A | 7 | 5.4 | codex-cli | MCP | Gives MCP-capable coding agents a local contract-lifecycle harness with governance audits and explicit completion gates. |
| openspec-skills (chyiiiiiiiiiiii) openspec-skills | A1 Skills-only | Tier A | 6 | 1 | claude-code | | Ports the OpenSpec proposal/apply/archive SDD lifecycle into three plain SKILL.md files requiring zero runtime dependencies. |
| vibe-stack (vibestackdev) vibe-stack | A8 Cross-runtime harness | Tier A | 6 | 1 | cursor | UI | Prevent AI coding assistants from generating insecure or broken Next.js 15 + Supabase code by injecting stack-specific constraint rules into every AI interaction. |
| pi-steering-hooks pi-steering-hooks | A11 Governance | Tier A | 5 | 2.1 | pi-coding-agent | | Pre-tool regex guardrails for the pi coding agent — deterministic, zero-token enforcement of behavioral rules with a self-authorizing override escape hatch. |
| cline-max (theogbrand) cline-max | A12 UI passthrough | Tier A | 4 | 2 | vscode | UI | VS Code extension fork that adds Plan/Act mode separation to enforce architecture-before-code discipline. |
| colign colign | A3 MCP-anchored | Tier A | 4 | 1.9 | claude-code | UI MCP | Provides a team-alignment platform for spec-driven development where teams discuss, write, and co-review structured proposals with BDD acceptance criteria before any AI agent writes code. |
| cursor-coding-agent-os (Mugiwara555343) cursor-coding-agent-os | A19 Token-efficient encoding | Tier A | 3 | 0.5 | cursor | | Lean/Verbose dual-mode Agent OS fork for solo developers on token budgets. |
| spec-kit-command-cursor (foxgod183) spec-kit-command-cursor-foxgod | A4 Markdown scaffold | Tier A | 3 | 0.5 | cursor | | Simplified Cursor IDE spec-driven toolkit with 8 commands and compact template variants for non-technical users. |
| AI Flywheel (agent-flywheel-plugin) ag-coding-flywheel | A11 Governance | Tier A | 2 | 7.1 | claude-code | MCP | Prevent multi-agent chaos through a structured bead lifecycle with file reservations, typed completion attestations, adversarial duel review, and 36-code structured error contracts. |
| cc-memory (skymanbp) cc-memory | A10 Memory engine | Tier A | 2 | 4 | claude-code | CLI UI MCP | SQLite memory with anti-patch upsert_smart (MERGE/SUPERSEDE/INSERT via trigram-Jaccard), forced PROGRESS.md handoff via <system-reminder>, PreCompact extraction, and plan tracking with plan-refiner/plan-guardian subagents. |
| doodledood/codex-workflow doodledood-codex-workflow | A4 Markdown scaffold | Tier A | 2 | 2 | openai-codex | | Full-lifecycle Codex CLI skill pack built around documented LLM limitations, with 8-type specialized review orchestration and dynamic requirements discovery. |
| meta-agent-teams (jbrahy) meta-agent-teams | A18 Self-evolving | Tier A | 2 | 4.2 | claude-code | | Build self-improving AI agent teams via a supervised training loop: specialist agents advise, a meta-agent evolves prompts based on human feedback, and an independent auditor prevents constitutional drift. |
| Simple Agent Harness bquast-simple-agent-harness | A4 Markdown scaffold | Tier A | 0 | 2.7 | none | | Minimal autonomous LLM agent: cron + Ollama + soul/memory/heartbeat markdown files, no dependencies. |
| Claude Conductor (MZWASHERE) claude-conductor-mzwashere | A6 Multi-agent orchestrator | Tier A | 0 | 5.9 | claude-code | | Exact copy of lackeyjb/claude-conductor with spam README; no independent value. |
| claudeAutoPilot claudeautopilot | A4 Markdown scaffold | Tier A | 0 | 4.9 | claude-code | UI | PRD-driven autonomous development framework that manages a 4-column Kanban workflow, runs quality gates, and surfaces only classified decisions (deploy/architecture/scope/security) to the human. |
| Flow (free framework) flow-free | A13 Methodology | Tier A | 0 | 0 | unknown | | Unknown — Reddit-only reference; no public implementation found. |
| roo-code-community (James-Cherished-Inc) james-cherished-roo-community | A4 Markdown scaffold | Tier A | 0 | 0 | roo-code | UI MCP | Community hub for Roo Code users: one-command custom mode installation, Perplexity MCP setup, and a web UI for managing .roomodes configurations. |
| MultiTable multitable | A13 Methodology | Tier A | 0 | 0 | unknown | | Unknown — Reddit-only reference; no public implementation found. |
| openspec-reviewed-workflow openspec-reviewed-workflow | A11 Governance | Tier A | 0 | 2.6 | cursor | | Inserts a mandatory evidence-driven codebase investigation gate between proposal and specs in OpenSpec workflows, forcing the AI to investigate existing solutions before specifying new ones. |
| saflib-workflows saflib-workflows | A13 Methodology | Tier A | 0 | 4.4 | claude-code | CLI | TypeScript-typed, composable workflow step machines that guide AI agents through structured multi-step development procedures with resume capability and optional per-step git commits. |
| spec-kit-aider (wolffy-au) spec-kit-aider | A4 Markdown scaffold | Tier A | 0 | 2.5 | aider | CLI | Aider-adapted fork of GitHub spec-kit that adds automated numbered git branch creation and cross-platform scripts for spec-driven development. |
| spec-ops-agents-file spec-ops-agents-file | A13 Methodology | Tier A | 0 | 1 | claude-code | | Embed spec-first discipline into any AI agent via AGENTS.md instructions: always read the spec first, never implement undocumented behavior, stop and request clarification when specs are ambiguous. |
| SpecSmith specsmith | A13 Methodology | Tier A | 0 | 1.3 | unknown | | Defines a minimal two-part spec contract (outcome + test gates) that compels verifiable thinking before agentic implementation, with zero tooling dependencies. |
| Agent Orchestrator (ComposioHQ) agent-orchestrator-composio | A7 PR-lifecycle platform | Tier A | — | 1.4 | unknown | | |
| DevTeam: Claude Code Multi-Agent Dev System devteam-michael-harris | A21 LLM-evaluated hooks | Tier A | — | 0 | unknown | | |
| DocBrain docbrain | A8 Cross-runtime harness | Tier A | — | 3.6 | unknown | CLI UI | |
| AI Agents in Enterprise Product Development Spec enterprise-ai-agents-spec | A4 Markdown scaffold | Tier A | — | 0 | unknown | | |
| Jeffallan/claude-skills jeffallan-claude-skills | A4 Markdown scaffold | Tier A | — | 0 | unknown | | |
| Maestro maestro-orchestrate | A8 Cross-runtime harness | Tier A | — | 1 | unknown | | |
| PAF Framework paf-framework | A4 Markdown scaffold | Tier A | — | 0 | unknown | | Repository not found — crack00r/paf-framework returns 404. |
| Patchwork OS patchwork-os | A8 Cross-runtime harness | Tier A | — | 1.4 | unknown | CLI | |
| roocode-memorybank-optimized roocode-memorybank-optimized | A4 Markdown scaffold | Tier A | — | 0 | unknown | | unknown — repository not accessible |
| squad-framework squad-framework | A4 Markdown scaffold | Tier A | — | 0 | unknown | | |
| oh-my-claudecode (Yeachan-Heo) oh-my-claudecode-yeachan | A16 Cross-vendor router | Tier B | 34,919 | 8.1 | claude-code | CLI MCP | Zero-learning-curve teams-first multi-agent orchestration for Claude Code with autopilot (6-phase lifecycle), ralph (PRD-driven persistence), ralplan (3-stage consensus), deep-interview (Socratic ambiguity gating), and tmux-based multi-CLI team workers. |
| NanoClaw nanoclaw | A8 Cross-runtime harness | Tier B | 29,431 | 4.8 | claude-code | CLI UI | Run Claude agents securely in per-session Docker containers with multi-channel messaging (WhatsApp/Telegram/Discord) and credential vault isolation. |
| SuperClaude Framework superclaude | A2 Mirror cmd+skill | Tier B | 23,009 | 5 | claude-code | CLI MCP | Meta-programming configuration framework that transforms Claude Code with 30 commands, 20 domain-specialist agents, and a PDCA self-improvement loop for systematic software development. |
| Planning with Files planning-with-files | A8 Cross-runtime harness | Tier B | 22,134 | 4 | claude-code | | Enforces Manus-style persistent markdown planning on any AI coding agent via hooks that automatically re-inject plan state before every tool call. |
| Ralph (snarktank) ralph-snarktank | A6 Multi-agent orchestrator | Tier B | 19,612 | 6.5 | amp | | Run an AI coding tool in a fresh-context loop, one user story at a time, until all stories in prd.json pass. |
| Personal AI Infrastructure (PAI) pai-personal-ai-infra | A20 OS-level substrate | Tier B | 14,434 | 6.6 | claude-code | UI | Life Operating System that drives every task as a current→ideal-state transition through a 7-phase Deutsch-epistemology Algorithm, with ISA specs, Pulse life dashboard, 45 skills, 18 agents, and file-based memory. |
| Aperant aperant | A6 Multi-agent orchestrator | Tier B | 14,304 | 2.9 | claude-code | UI | Cross-platform Electron desktop for autonomous multi-agent coding with kanban UI, parallel git worktrees, self-validating QA loop, and GitHub/Linear integration. |
| IronClaw ironclaw | A9 Sandbox substrate | Tier B | 12,351 | 3.3 | ironclaw-runtime | CLI UI MCP | WASM-sandboxed personal AI runtime with defense-in-depth security: capability-based tool isolation, credential vault, prompt injection defense, and endpoint allowlisting. |
| HumanLayer / CodeLayer humanlayer | A11 Governance | Tier B | 10,896 | 4.7 | claude-code | CLI UI MCP | Orchestrate parallel Claude Code sessions per Linear ticket via dedicated git worktrees (MULTICLAUDE pattern) with structured research→plan→implement→PR workflows and a Tauri desktop app for session management. |
| Ralph for Claude Code ralph-claude-code | A6 Multi-agent orchestrator | Tier B | 9,205 | 4.9 | claude-code | CLI UI | Production-grade Claude Code autonomous loop with dual-condition exit gate, rate limiting, circuit breaker, session management, and live monitoring. |
| Trellis trellis-mindfold | A21 LLM-evaluated hooks | Tier B | 8,467 | 5.4 | claude-code | CLI | Team-scale AI coding harness: shared specs auto-injected via hooks into sub-agents across 14 AI tools, with per-developer workspace journals and spec-learning loop. |
| Claude Code PM (ccpm) ccpm | A8 Cross-runtime harness | Tier B | 8,146 | 4.1 | claude-code | | Five-phase PRD-to-shipped-code skill using GitHub Issues as canonical task store with parallel agent execution via git worktrees. |
| Paseo paseo | A16 Cross-vendor router | Tier B | 6,754 | 3.5 | claude-code | CLI UI MCP | Multi-provider AI coding agent orchestration daemon with cross-device access (phone/desktop/CLI) and git worktree isolation. |
| ZCF — Zero-Config Code Flow zcf | A6 Multi-agent orchestrator | Tier B | 6,009 | 2.5 | claude-code | CLI | Zero-config one-command setup for Claude Code + BMAD methodology + API relay provider configuration. |
| Backlog.md backlog-md | A8 Cross-runtime harness | Tier B | 5,622 | 1.5 | claude-code | CLI UI MCP | Structured task management for AI coding agents using per-task markdown files, MCP, and a web Kanban board — all inside the git repo. |
| OSAURUS osaurus | A12 UI passthrough | Tier B | 5,509 | 4.7 | osaurus-native | CLI UI MCP | Native macOS AI harness with Apple Container Linux VM sandboxing, on-device privacy filter, MLX inference, and cryptographic identity. |
| CCG Workflow ccg-workflow | A16 Cross-vendor router | Tier B | 5,357 | 7.6 | claude-code | CLI | Routes Claude + Codex + Gemini to task-appropriate collaboration strategies (direct-fix through full-collaborate) with hook-based per-turn state tracking that survives context compaction. |
| HiClaw hiclaw | A6 Multi-agent orchestrator | Tier B | 4,666 | 6.1 | kubernetes | CLI UI | Kubernetes-native multi-agent platform where Worker and Manager agents collaborate in a Matrix IM room with enterprise credential isolation via Higress AI gateway and shared MinIO storage. |
| GitHub Agentic Workflows gh-aw-githubnext | A11 Governance | Tier B | 4,525 | 7.5 | github-copilot | CLI | Write agentic workflows in natural language markdown, compile to GitHub Actions, run with production-grade security guarantees — read-only permissions by default, all writes through sanitized safe-outputs. |
| Continuous Claude continuous-claude-parcadei | A7 PR-lifecycle platform | Tier B | 3,789 | 8.1 | claude-code | | Compound learning across Claude Code sessions via PostgreSQL memory extraction from thinking blocks, YAML handoffs for session transfer, 32 specialized agents, and 109 skills with natural-language activation. |
| Atmosphere atmosphere | A21 LLM-evaluated hooks | Tier B | 3,767 | 7.5 | spring-ai | CLI UI MCP | JVM framework for building production-grade governed AI agents with streaming transports, HITL approvals, Cedar/Rego policy admission, and durable workflow checkpoints. |
| TinyAGI tinyagi | A6 Multi-agent orchestrator | Tier B | 3,568 | 4.5 | claude-code | CLI UI | 24/7 multi-agent multi-team platform with Discord/Telegram/WhatsApp channels, peer-to-peer message-passing coordination, and a Next.js web portal for one-person-company workflows. |
| CCPlugins (brennercruvinel) ccplugins-cruvinel | A4 Markdown scaffold | Tier B | 2,705 | 0.9 | claude-code | | 24 professional Claude Code commands that prevent AI overengineering and automate repetitive development tasks. |
| AI-DLC Workflows (AWS) aws-aidlc-workflows | A8 Cross-runtime harness | Tier B | 2,443 | 5.4 | kiro | | Cross-IDE adaptive software development lifecycle enforcement for AI coding agents, with mandatory audit trails and human approval gates at each phase. |
| centminmod/my-claude-code-setup centminmod-cc-setup | A8 Cross-runtime harness | Tier B | 2,359 | 4.2 | claude-code | MCP | Provides a memory-resilient Claude Code starter with dual git-shared + machine-local memory that survives CLAUDE.md resets, plus multi-provider delegation to Codex and Z.AI. |
| MCP Agent Mail mcp-agent-mail | A6 Multi-agent orchestrator | Tier B | 1,958 | 4.9 | claude-code | CLI UI MCP | Provides an asynchronous mail-like coordination bus (identities, inboxes, file leases, Git-auditable messages) so multiple coding agents can work the same codebase in parallel without stepping on each other. |
| claude-codepro (Pilot Shell) claude-codepro | A6 Multi-agent orchestrator | Tier B | 1,728 | 8.5 | claude-code | CLI UI MCP | Adds spec-driven planning, enforced TDD, persistent memory, multi-agent review, and a local web dashboard to Claude Code — making it production-grade for engineering teams. |
| claude-code-settings (feiskyer) feiskyer-cc-settings | A16 Cross-vendor router | Tier B | 1,525 | 3.9 | claude-code | UI MCP | Provide a multi-model Claude Code setup routed through LiteLLM + GitHub Copilot, with skills for deep research, cross-tool delegation, and code workflows. |
| codeg codeg | A12 UI passthrough | Tier B | 1,344 | 5.2 | desktop-app | CLI UI | Unifies multiple heterogeneous AI coding agents into one desktop workspace with session aggregation and cross-agent delegation. |
| Context Engineering Kit context-engineering-kit-neolabhq | A19 Token-efficient encoding | Tier B | 1,042 | 1.4 | claude-code | | Token-efficient multi-plugin engineering kit with automatic adversarial self-reflection and granular plugin installs. |
| Optio optio | A3 MCP-anchored | Tier B | 967 | 5.9 | claude-code | UI MCP | Self-hosted Kubernetes-native AI engineering platform that orchestrates ticket-to-PR automation with automatic CI/review feedback loops, multi-vendor agent support, and persistent inter-agent messaging — code never leaves your infrastructure. |
| Tsumiki tsumiki | A6 Multi-agent orchestrator | Tier B | 965 | 7.6 | claude-code | | Japanese-developed Claude Code plugin delivering 40+ commands covering requirements (EARS) → design → TDD → implementation → verification for full project lifecycle. |
| Blueprint (JuliusBrussee) / cavekit blueprint-juliusbrussee | A19 Token-efficient encoding | Tier B | 948 | 5.4 | claude-code | | Compresses the entire SDD workflow into a single SPEC.md file using caveman encoding (~75% fewer tokens) with automatic bug-to-invariant backpropagation on test failure. |
| Loki Mode loki-mode | A6 Multi-agent orchestrator | Tier B | 944 | 10 | claude-code | CLI UI MCP | Takes a spec (PRD/issue/OpenAPI) and autonomously produces a production-ready Git repository through RARV cycles, 11 blocking quality gates, and 41 specialized agents. |
| spec_driven_develop zhu1090093659-sdd | A8 Cross-runtime harness | Tier B | 866 | 7.5 | claude-code | | 6-phase cybernetics-inspired pipeline for large-scale AI transformations with S.U.P.E.R architectural health framework, adaptive drift control, and GitHub-native task tracking. |
| 10x-Tool-Calls 10x-tool-calls | A4 Markdown scaffold | Tier B | 863 | 0.5 | cursor | | Maximize the number of tasks accomplished within a single Cursor/Windsurf request by running a user-input loop between tasks. |
| GSD for Antigravity gsd-for-antigravity | A10 Memory engine | Tier B | 862 | 4.1 | claude-code | | Make AI coding reliable through context engineering: structured SPEC→PLAN→EXECUTE→VERIFY loop with wave-based atomicity and model-agnostic portability. |
| evo evo | A6 Multi-agent orchestrator | Tier B | 770 | 6.4 | claude-code | CLI UI | Autonomously optimize code through parallel tree-search experiments with shared state, gate-validated results, and configurable frontier strategies. |
| Corsair corsair | A3 MCP-anchored | Tier B | 759 | 2.7 | claude-code | UI MCP | Unified permissioned integration layer for AI agents: 70+ third-party services accessible via typed API calls with credential isolation and human-approval gates for destructive actions. |
| Claude Bootstrap + Maggy claude-bootstrap | A16 Cross-vendor router | Tier B | 670 | 8.6 | claude-code | CLI UI MCP | Route every task to the cheapest capable model while enforcing TDD, ADR compliance, iCPG intent tracking, and Telos intent fidelity — making token economics a first-class engineering discipline. |
| Clavix clavix | A4 Markdown scaffold | Tier B | 656 | 2.3 | claude-code | CLI | Enforces strict PRD/implementation phase separation via mandatory State Assertions and a 6-type Self-Correction Protocol across 20+ AI coding tools. |
| power-bi-agentic-development power-bi-agentic-dev | A6 Multi-agent orchestrator | Tier B | 619 | 0.4 | claude-code | MCP | Makes Claude Code a competent Power BI/Fabric developer by providing 25 domain skills, 8 validator agents, and PostToolUse hooks that enforce TMDL/PBIR/DAX format correctness automatically. |
| altimate-code altimate-code | A2 Mirror cmd+skill | Tier B | 604 | 5.2 | standalone | CLI UI | A full agentic data engineering harness with deterministic SQL intelligence (100% F1 anti-pattern detection, column-level lineage, PII detection) for dbt, SQL, and 10+ cloud warehouses. |
| Clean-Coder-AI clean-coder-ai | A6 Multi-agent orchestrator | Tier B | 576 | 6.9 | standalone-python | | Autonomous 2-in-1 AI developer + scrum master that plans an entire project as Todoist tasks and executes them task-by-task with a LangGraph multi-agent pipeline. |
| oh-my-opencode (opensoft) opensoft-oh-my-opencode | A8 Cross-runtime harness | Tier B | 544 | 4.5 | opencode | CLI MCP | Multi-provider multi-agent harness for OpenCode with Sisyphus orchestrator routing work across Anthropic/OpenAI/Google/xAI agents based on task domain. |
| claude-code-bmad-skills claude-code-bmad-skills | A19 Token-efficient encoding | Tier B | 435 | 2.7 | claude-code | | Token-efficient re-implementation of BMAD-METHOD using Claude Code's native skill-md format with 3 lifecycle hooks and a project-level taxonomy for workflow routing. |
| notque/claude-code-toolkit notque-cc-toolkit | A11 Governance | Tier B | 388 | 4.8 | claude-code | UI | Enterprise-scale Claude Code toolkit with 44 domain agents, 77 Python hooks, a confidence-decaying SQLite learning database, AFK headless mode, and formal prompt injection defense. |
| VibeDoc vibedoc | A12 UI passthrough | Tier B | 373 | 0 | claude-code | UI | Generate structured development plans from natural language descriptions in 60-180 seconds via a Gradio web UI, exportable to MD/Word/PDF/HTML for handoff to any AI coding tool. |
| Claude Conductor (superbasicstudio fork) claude-conductor-superbasic | A4 Markdown scaffold | Tier B | 367 | 3.5 | claude-code | CLI | Gives Claude Code a persistent, cross-linked, auto-analyzed documentation system so it retains codebase context across sessions. |
| second-brain second-brain-nicholas | A10 Memory engine | Tier B | 355 | 2.2 | claude-code | UI | LLM-maintained personal knowledge base: drop raw sources into a folder, let the LLM compile them into a structured Obsidian wiki with entities, concepts, and synthesis pages. |
| GolemBot golembot | A16 Cross-vendor router | Tier B | 285 | 2.5 | claude-code | CLI UI | Bridges any coding agent (Claude Code, Cursor, Codex, OpenCode) to IM channels and HTTP APIs, giving agents external reach without framework lock-in. |
| OpenASE openase | A6 Multi-agent orchestrator | Tier B | 243 | 7 | openase binary | CLI UI | Turns tickets into working code via AI agents that autonomously claim tickets, execute versioned workflow harnesses, and pass lifecycle hook quality gates. |
| BMAD-AT-CLAUDE bmad-at-claude | A8 Cross-runtime harness | Tier B | 235 | 4 | claude-code | CLI | Ports the full BMAD agile agent workflow (10 named personas, planning+implementation phases) into Claude Code's native hook and subagent infrastructure. |
| Liza liza | A17 Compiled enforcement | Tier B | 227 | 8.5 | claude-code | CLI UI | Hardened multi-agent coding system with code-enforced role boundaries, adversarial doer/reviewer pairs, and 55+ failure mode countermeasures — turning agent behavioral trust from a prompt engineering problem into a compiled Go enforcement problem. |
| Switchboard switchboard | A16 Cross-vendor router | Tier B | 209 | 3.2 | vscode-extension | UI MCP | Orchestrate heterogeneous AI agent teams via visual kanban drag-and-drop, routing tasks by complexity to the right model without writing prompts. |
| Shep shep-cli | A8 Cross-runtime harness | Tier B | 194 | 6.5 | shep CLI | CLI UI | Runs multiple AI coding agents in parallel, each in its own git worktree, handling commits, PRs, and CI monitoring automatically. |
| alexeykrol/claude-code-starter v6 alexeykrol-cc-starter | A6 Multi-agent orchestrator | Tier B | 180 | 3.5 | claude-code | | Russian-language Claude Code starter v6 with auto project-type detection, SNAPSHOT.md session state, pre/post compact hooks, and repo_access git-history control. |
| oh-my-claude (TechDufus) techdufus-oh-my-claude | A6 Multi-agent orchestrator | Tier B | 147 | 6.5 | claude-code | | Adds three-reviewer quality gate chain (advisor+critic+risk-assessor) and Python-hook enforcement layer to Claude Code before execution. |
| serpro69/claude-toolbox serpro69-claude-toolbox | A16 Cross-vendor router | Tier B | 142 | 2.9 | claude-code | MCP | Structured 11-skill development pipeline (design→review→implement→review→test→doc) with persistent Capy BM25 knowledge base, language-specific profiles for 8 languages, and Go-based cross-provider codegen targeting both Claude Code and Codex from a single canonical source. |
| OpenLore (formerly spec-gen) spec-gen | A19 Token-efficient encoding | Tier B | 137 | 3.7 | claude-code | CLI UI MCP | Give AI agents persistent architectural orientation by reverse-engineering OpenSpec living specs from code and exposing the graph via 45 MCP tools, reducing per-session orientation cost from 30k tokens to 1k. |
| Autospec autospec | A6 Multi-agent orchestrator | Tier B | 130 | 7.4 | claude-code | CLI | Go CLI that orchestrates spec → plan → tasks → implement in fresh isolated agent sessions with YAML artifacts and programmatic validation. |
| spec-based-claude-code spec-based-claude-code | A4 Markdown scaffold | Tier B | 129 | 4.5 | claude-code | | Ten slash commands enforce requirements→design→tasks→implement with file-marker approval gates in plain Claude Code, zero dependencies. |
| cloudnative-co/claude-code-starter-kit cloudnative-cc-starter | A6 Multi-agent orchestrator | Tier B | 116 | 1.9 | claude-code | | Japanese-language wizard-driven Claude Code setup kit with 3 profiles, Anthropic plugin marketplace integration, and cross-platform (macOS/Linux/Windows) installation. |
| opencode-plugin-openspec opcode-plugin-openspec | A4 Markdown scaffold | Tier B | 109 | 2 | opencode | | Prevent premature code implementation during OpenSpec planning by injecting a write-restricted planning agent into OpenCode. |
| AiDE aide | A4 Markdown scaffold | Tier B | 106 | 1.7 | claude-code | | Standardizes AI context management across sessions with a zero-dependency markdown folder convention. |
| Shadow Code shadow-code | A5 Closed IDE | Tier B | 79 | 0.5 | vscode | UI | Developers write pseudocode in a parallel .shadow file; Shadow Code transforms it into target-language production code via VS Code's Language Models API, producing more deterministic output than vibe coding. |
| OpenSpecUI (jixoai) openspec-ui-jixoai | A8 Cross-runtime harness | Tier B | 72 | 1.8 | claude-code | CLI UI | Provide a visual web interface for OpenSpec workflows with PTY terminal, OPSX compose, project hooks, and static export. |
| ORCH orch-oxgeneral | A7 PR-lifecycle platform | Tier B | 68 | 8 | claude-code | CLI UI | Orchestrates teams of parallel AI agents (CTO + workers + reviewer) on a single codebase with YAML flat-file state, a TUI dashboard, and CI-friendly headless mode. |
| Blueprint (Imbue) blueprint-imbue | A1 Skills-only | Tier B | 63 | 2.3 | claude-code | | Slows down AI coding just enough to ask codebase-grounded clarifying questions before writing the implementation plan, with user-controlled Q&A termination. |
| SDD (Solokit) sdd-session-ankushdixit | A4 Markdown scaffold | Tier B | 61 | 3.7 | claude-code | CLI | Eliminate AI context loss between sessions by reconstructing full work state from git history and spec files at each session start. |
| Aegis aegis | A4 Markdown scaffold | Tier B | 60 | 3.5 | claude-code | | Gives AI assistants a shared project state machine so sessions don't lose context and tasks have verifiable state transitions. |
| agentic-agile-template (Microsoft) agentic-agile-template-ms | A4 Markdown scaffold | Tier B | 60 | 3.7 | github-copilot | | A GitHub template providing the CLAUDE.md, Issue templates, manifesto, and measurement framework teams need to adopt wave-based parallel human-agent development. |
| Octomind octomind-muvon | A16 Cross-vendor router | Tier B | 59 | 5.2 | openrouter | CLI UI | Provides a specialist AI agent runtime with tap-registry community agents, per-role multi-model routing, hard cost enforcement, and policy-as-code guardrails — all in a single Rust binary. |
| FredAntB/Spec-Driven-Development spec-driven-fredantb | A8 Cross-runtime harness | Tier B | 57 | 4 | claude-code | | Conversational skill that interviews the user once, generates requirements/design/tasks, then creates identical Universal Instruction Blocks for every AI tool so they can't contradict each other. |
| Asynkor asynkor | A3 MCP-anchored | Tier B | 49 | 5 | any MCP agent | UI MCP | Prevents edit-time file conflicts and provides compounding team memory for concurrent multi-machine AI agent teams. |
| ClawSpec clawspec | A6 Multi-agent orchestrator | Tier B | 43 | 4 | openclaw | | Embed OpenSpec workflow into OpenClaw chat with visible planning, durable background implementation that survives gateway restarts, and channel-scoped project state. |
| openspec-schemas (intent-driven-dev) openspec-schemas-intent | A8 Cross-runtime harness | Tier B | 43 | 2.2 | unknown | | Extends OpenSpec with domain-specific artifact workflows for BDD, event-driven systems, ADRs, and Linear project management. |
| Dev Workspace dev-workspace-dilberry | A10 Memory engine | Tier B | 39 | 4.2 | claude-code | | Git-branch-scoped context management system that isolates planning, research, history, and context per branch while providing automatic conversation capture with headless-Claude summarization. |
| claude-code-settings-for-unity nowsprinting-cc-unity | A11 Governance | Tier B | 37 | 3 | claude-code | | Enforces test-first Unity C# development via a two-agent (Plan + test-designer) pipeline with a TESTABILITY gate before any code is written. |
| Vibe Architect vibe-architect | A13 Methodology | Tier B | 34 | 2 | claude-code | UI | Convert a raw app idea into a coding-agent-ready implementation spec through guided multi-phase conversation with live visual design previews. |
| Personal OS (itseffi) agentic-os-effi | A4 Markdown scaffold | Tier B | 33 | 1.5 | claude-code | MCP | Provides a file-based personal operating system that grounds AI agent execution in explicit goals, backlogs, and 47 cross-domain skills spanning engineering, product strategy, and stakeholder management. |
| ai-ticket ai-ticket-jmike | A4 Markdown scaffold | Tier B | 33 | 3.2 | unknown | | Human-as-LLM-proxy ticket system where each AI-human interaction step creates a GitHub Issue ticket; designed for AutoGPT-era human-in-the-loop workflows. |
| AI Development Specifications (cogeet-io) cogeet-ai-dev-specs | A13 Methodology | Tier B | 32 | 0 | claude-code | | Apply Infrastructure-as-Code principles to AI development processes — spec-as-code, testing-as-code, context-engineering-as-code — for systematic, measurable AI-assisted development. |
| pbip-demo-agentic-mcp pbip-demo-agentic-mcp | A13 Methodology | Tier B | 30 | 1 | github-copilot | MCP | Demonstrates spec-first Power BI semantic model development using powerbi-modeling-mcp with BPA-enforced quality loops. |
| spec-first (sunrain520) spec-first-sunrain | A6 Multi-agent orchestrator | Tier B | 30 | 6.6 | claude-code | CLI | npm CLI + 40 skills + 51 agent reviewers that turns AI coding into a repeatable loop where scripts prepare deterministic facts and LLMs make semantic decisions. |
| novel-writer-openspec novel-writer-openspec | A4 Markdown scaffold | Tier B | 29 | 2.2 | unknown | CLI | Applies OpenSpec methodology to fiction writing to prevent character collapse and setting inconsistency in AI-assisted long-form novels. |
| Claude for Reservoir Engineering reservoir-engineering-claude | A13 Methodology | Tier B | 28 | 1 | claude-code | MCP | Teaches petroleum engineers to use Claude Code through domain-authentic exercises, with a reviewer subagent that checks unit consistency, correlation applicability, and pyResToolbox parameter naming. |
| Deliberate Agentic Development deliberate-agentic-dev | A4 Markdown scaffold | Tier B | 26 | 5.5 | claude-code | MCP | Structured AI agent workflow with Linear integration, human-review gates at every commit/PR/milestone, and state.json position tracking for precise resume. |
| Flokay (Codagent Agent Skills) flokay | A7 PR-lifecycle platform | Tier B | 26 | 5.8 | Claude Code | | Full SDLC skill pack from idea evaluation through CI-green PR, with TDD Iron Law enforcement and external validator quality gates. |
| claude-code-pro-pack sisyphusse-pro-pack | A18 Self-evolving | Tier B | 26 | 0.5 | claude-code | UI | Provides a 12-rule drop-in CLAUDE.md behavior baseline (building on Karpathy's 4-rule original) that addresses agent-orchestration failures: token spirals, silent failures, two-pattern pollution, duplicate drift. |
| spec-driver (davidlee) spec-driver-davidlee | A4 Markdown scaffold | Tier B | 25 | 6 | claude-code | CLI UI | Python CLI + 24 skills + TUI for evergreen specification management — specs that stay accurate as implementations evolve, not disposable planning documents. |
| Indie Kit ind-kit | A8 Cross-runtime harness | Tier B | 22 | 0.9 | claude-code | UI | A Next.js 16 SaaS starter kit with embedded Claude Code skills that prevent AI agents from hallucinating incompatible patterns by encoding exact import paths, API conventions, and antipatterns for the project's specific tech stack. |
| matsengrp/plugins matsengrp-plugins | A6 Multi-agent orchestrator | Tier B | 22 | 0.4 | claude-code | | Gives computational biology teams 12 specialized agents for scientific writing, LaTeX editing, bioinformatics pipelines, and code quality review, with macOS notification hooks. |
| AI Task Master (James-Cherished-Inc) ai-task-master-james | A4 Markdown scaffold | Tier B | 21 | 3.3 | roo-code | CLI | Fork of claude-task-master adapted for Roo Code, Cline, and OpenRouter multi-model API routing. |
| Spek spek | A12 UI passthrough | Tier B | 21 | 0 | vscode | UI | Give developers a read-only, multi-IDE viewer for OpenSpec content with git worktree aggregation for parallel AI agent workflows. |
| Tracer tracer-issue-tracker | A8 Cross-runtime harness | Tier B | 20 | 1.7 | claude-code | CLI | Dependency-aware JSONL issue tracker for AI agents with a ready-queue that surfaces only unblocked work. |
| research-kit (nguyenvanduocit) research-kit-nguyenvanduocit | A6 Multi-agent orchestrator | Tier B | 19 | 3.6 | claude-code | CLI | Adapts the spec-kit SDD workflow for systematic 10-phase AI-guided research with source provenance tracking and BibTeX citations. |
| Harness-Engineering-skills harness-engineering-skills-phlegon | A6 Multi-agent orchestrator | Tier B | 17 | 5.9 | claude-code | CLI | PRD-to-code repo-backed delivery orchestration with three ceremony levels, 19 named agents, filesystem-enforced phase gates, and cross-session state persistence. |
| ZCC — oh-my-zsh for Claude Code zcc | A10 Memory engine | Tier B | 17 | 1 | claude-code | CLI | Bring oh-my-zsh concepts (themes as modes, plugins as packs, tmux sessions as tickets) to Claude Code for persistent, context-aware development. |
| Polka Codes polka-codes | A6 Multi-agent orchestrator | Tier B | 16 | 5.9 | standalone-cli | CLI MCP | TypeScript CLI that provides composable AI coding subcommands with multi-provider routing, checkpoint/resume, and a TypeScript API for meta-workflow scripting. |
| SprintCore sprintcore | A10 Memory engine | Tier B | 16 | 1 | unknown | CLI | CLI tool converting PRD markdown into structured Epics/Stories/Subtasks, plus a FAISS-indexed codebase bug-fix agent. |
| bmad-architecture-agent bmad-architecture-agent | A11 Governance | Tier B | 15 | 2.2 | claude-code | | Provides 8 specialized architecture expert AI personas (cloud, data, integration, platform, governance) as a BMAD expansion pack for architecture-heavy projects. |
| zbruhnke/claude-code-starter zbruhnke-cc-starter | A6 Multi-agent orchestrator | Tier B | 15 | 2.4 | claude-code | CLI | Prevents vibe-coding via a Wiggum Loop autonomous implementation skill with OS-level git pre-commit enforcement of test/lint/changelog gates. |
| openspec-spec-driven-superpowers openspec-spec-driven-superpowers | A8 Cross-runtime harness | Tier B | 14 | 6 | claude-code | | Bridges OpenSpec's lifecycle control with superpowers execution discipline via a review.md readiness gate and per-change execution mode selection. |
| alex-feel/claude-code-toolbox alex-feel-cc-toolbox | A4 Markdown scaffold | Tier B | 13 | 1.5 | claude-code | | YAML-declarative meta-installer for Claude Code environments: define agents, MCP servers, hooks (command/HTTP/prompt/agent), model settings, and permissions in one YAML file and install cross-platform with one command. |
| my-claude-code-harness (code-yeongyu) code-yeongyu-my-cc-harness | A6 Multi-agent orchestrator | Tier B | 13 | 3 | claude-code | | Personal Claude Code workflow: plan-to-execute pipeline with Haiku workers, profane attention-weight prompting for atomic task enforcement, and PostToolUse Python static analysis hooks. |
| pi-ralph pi-ralph-orch | A6 Multi-agent orchestrator | Tier B | 13 | 6 | pi-coding-agent | UI | Add YAML-configurable hat-based multi-phase orchestration loops to the pi coding agent, with event-driven hat transitions and guard rails. |
| Claude-Slash-Commands (danielrosehill) claude-slash-commands-daniel | A4 Markdown scaffold | Tier B | 10 | 0.6 | claude-code | | Provide a large library of reusable slash commands across 25 development domains for Claude Code. |
| CC Command Center / Local Qwen Advisor cc-command-center | A16 Cross-vendor router | Tier B | 9 | 3.1 | unknown | UI | Passive session observability (FTS5 dashboard over JSONL) + inter-LLM delegation (Claude calling Qwen as a cheaper advisory model) |
| Prodigy prodigy-iepathos | A6 Multi-agent orchestrator | Tier B | 9 | 5.9 | claude-code | CLI | Turn ad-hoc Claude sessions into reproducible, parallel, checkpointable development pipelines via YAML workflow files. |
| CueAPI cueapi | A11 Governance | Tier B | 8 | 4.4 | REST API | | Replaces cron for AI agent tasks with outcome-aware scheduling: success/failure tracking, automatic retries, and execution history. |
| oh-my-claude (stefandevo) stefandevo-oh-my-claude | A6 Multi-agent orchestrator | Tier B | 7 | 5.3 | claude-code | MCP | Ports oh-my-opencode's Sisyphus multi-agent orchestration to Claude Code natively after Anthropic blocked third-party OAuth access. |
| DotForge dotforge | A17 Compiled enforcement | Tier B | 6 | 4 | claude-code | | Declare behavioral policies for Claude Code and compile them into enforcing PreToolUse hooks, with cross-project audit and sync governance. |
| Sparo OS sparo-os | A20 OS-level substrate | Tier B | 6 | 3 | sparo-os | UI | Cross-platform Tauri desktop application that acts as an Agentic OS runtime — hosting continuous AI task execution, workspace management, and dynamically-generated Live Apps across Code, Cowork, and Design built-in apps with mobile remote control. |
| wyndomb/claude-code-starter wyndomb-cc-starter | A4 Markdown scaffold | Tier B | 6 | 0 | claude-code | MCP | A productivity-workflow Claude Code starter for knowledge workers shipping daily planning, research, standup, and weekly synthesis commands with pre-wired Perplexity + Notion + Obsidian MCP. |
| AndAI andai | A4 Markdown scaffold | Tier B | 5 | 5.9 | unknown | CLI UI | Self-hosted AI development loop using Redmine + Aider + YAML-configurable workflows with multi-model routing and human review gates. |
| oh-my-claudecode (mazenyassergithub) oh-my-claudecode-mazen | A99 Unclassified | Tier B | 5 | 0 | claude-code | | Claims multi-agent orchestration with 28 agents and 28 skills but README content is SEO-style with ZIP download instructions inconsistent with Claude Code plugin architecture. |
| AI Dev Workflow Kit ai-dev-workflow-kit-aahil | A11 Governance | Tier B | 4 | 2.4 | unknown | | Portable .ai-kit/ folder enforcing EPCC workflow loop, Max-2 Resource Rule, and Decision Gates to prevent context overload and unplanned agent behavior. |
| OpenSpec Badge Action openspec-badge-action | A4 Markdown scaffold | Tier B | 4 | 1 | github-actions | | Generate OpenSpec health metrics as SVG badges for embedding in GitHub READMEs via CI. |
| openspec-playwright openspec-playwright | A3 MCP-anchored | Tier B | 4 | 4.7 | claude-code | CLI MCP | Extends OpenSpec's spec lifecycle to automated E2E verification by deriving Playwright tests directly from spec files via DOM exploration. |
| Specboard specboard | A12 UI passthrough | Tier B | 4 | 0 | openspec | CLI UI | Give human operators a real-time kanban dashboard over parallel OpenSpec worktrees without interfering with agent execution. |
| vibecodex vibecodex | A4 Markdown scaffold | Tier B | 4 | 2.5 | claude-code | CLI MCP | 54 numbered architecture rules (A1-F10) for FastAPI + Next.js + Go that AI coding agents follow automatically — enforcing layer boundaries, file-size limits, idempotency, and bulkhead isolation in vibe-coded projects. |
| oh-my-customcodex (baekenough) baekenough-oh-my-customcodex | A6 Multi-agent orchestrator | Tier B | 3 | 4 | codex | CLI | Codex-native port of oh-my-customcode: 49 agents, 123 skills, compiled from source using the skills-as-source-agents-as-artifacts metaphor, with dynamic agent creation for missing domains. |
| ACT-House-Engineering/codex-starter codex-starter-acthouse | A13 Methodology | Tier B | 3 | 1.4 | openai-codex | | Minimal OpenAI Codex starter with a container setup script, AGENTS.md behavioral guide, and reference docs for code quality workflows (twelve-factor, pragmatic programmer). |
| Compose-Lang compose-lang | A15 Harness generator | Tier B | 3 | 1.5 | compose CLI | CLI UI | Turns minimal architecture specs into production code via a compiler pipeline that calls an LLM backend. |
| HarnessOS harness-os | A8 Cross-runtime harness | Tier B | 3 | 5.9 | unknown | CLI MCP | SQLite-backed execution harness that gives AI agents task lifecycle management, lease-based concurrency control, workload-profile skill specialization, and Symphony-style four-agent fan-out with deterministic evidence gates. |
| maverick maverick-get2knowio | A6 Multi-agent orchestrator | Tier B | 3 | 6.4 | claude-code | CLI | Orchestrates PRD-to-merged-code via a bead dependency graph where agents and humans create work for each other, with enriched review context (spec + contrarian briefing + runway history). |
| OpenSpec.sh openspec-sh | A4 Markdown scaffold | Tier B | 3 | 0 | unknown | CLI | POSIX bash reimplementation of OpenSpec for environments without Node.js — dependency-free spec-driven workflow scaffold. |
| foomakers/pair ozzeron-foomakers-pair | A8 Cross-runtime harness | Tier B | 3 | 3.2 | claude-code | CLI | Gives AI assistants the full team context — PRD, ADRs, guidelines, process — needed to execute a complete SDLC without context loss. |
| agent-plugins-skills (richfrem) richfrem-agent-plugins-skills | A18 Self-evolving | Tier B | 3 | 5.7 | claude-code | UI MCP | Provide a self-improving, cross-platform plugin ecosystem where skills evolve through eval-gated iteration rather than human-written intuition. |
| sourcebook sourcebook | A13 Methodology | Tier B | 3 | 1.4 | claude-code | CLI MCP | Generates AI context files from codebase git history and import graph, capturing hidden co-change couplings and hub file blast radius that agents consistently miss. |
| SpecCrew speccrew | A4 Markdown scaffold | Tier B | 3 | 5.3 | claude-code | CLI | Apply ISA-95 industrial methodology to AI-assisted software delivery — 8 specialized agents orchestrate PRD through deployment with a 3-tier knowledge base and per-requirement iteration isolation. |
| cestDone cestdone | A16 Cross-vendor router | Tier B | 2 | 6.9 | claude-code | CLI | Automate Director+Worker AI development by keeping the Director context-thin and giving each phase Worker a fresh session with only relevant context. |
| claude-commands (oddballteam) claude-commands-oddball | A4 Markdown scaffold | Tier B | 2 | 0.9 | claude-code | | Provide analysis-only AI commands for engineering workflows with explicit no-automation policy for compliance-sensitive environments. |
| intellij-openspec intellij-openspec | A8 Cross-runtime harness | Tier B | 2 | 3.2 | intellij-idea | UI | Provide full OpenSpec lifecycle orchestration within IntelliJ IDEA with flexible AI routing to any of 28 detected tools or direct API providers. |
| nl-spec-driven nl-spec-driven-elliotone | A13 Methodology | Tier B | 2 | 0.5 | claude-code | | Eliminate implicit AI coding decisions by making scope, constraints, and task boundaries explicit in versioned markdown files with human approval gates. |
| OpenSpec-Video openspec-video | A6 Multi-agent orchestrator | Tier B | 2 | 6.4 | unknown | CLI UI | Spec-driven AI media production: compile narrative Markdown specs into production-ready images/videos via multi-provider pipeline with dependency graph management. |
| PlanGate plangate | A16 Cross-vendor router | Tier B | 2 | 10 | claude-code | CLI | Enforce a gated spec-driven AI development workflow with human approval at planning and PR stages, multi-model review, and full auditability via event logs. |
| spec-first (nlatuan187) spec-first-nlatuan | A13 Methodology | Tier B | 2 | 6.4 | claude-code | | Single-file paste-in methodology that engineers AI behavior via probability theory, with a Scope Routing table that maps error-state count to implementation route. |
| AI Dev Workflow (tkhongsap) ai-dev-workflow-tkhongsap | A6 Multi-agent orchestrator | Tier B | 1 | 2.6 | claude-code | | PRD-driven development workflow + multi-specialist code review system with GitLab MR integration, designed for enterprise teams mentoring junior developers. |
| OmniDev Kit omnidev-kit | A4 Markdown scaffold | Tier B | 1 | 6.4 | claude-code | | Full-lifecycle AI development workflow skill for Claude Code: safety-first requirement alignment, adaptive phase planning, self-learning from session retrospectives. |
| openspec-schemas (kmhalvin) openspec-schemas-kmhalvin | A8 Cross-runtime harness | Tier B | 1 | 5.5 | claude-code | | Prevents context budget exhaustion and hallucination on complex brownfield changes via subagent delegation and mandatory session restarts. |
| opsx-feature-dev opsx-feature-dev | A6 Multi-agent orchestrator | Tier B | 1 | 4.9 | Claude Code | | Adds a 7-phase artifact-driven feature development workflow on top of the OpenSpec CLI and Anthropic's official feature-dev plugin. |
| Agentic OS (KbWen) agentic-os-kbwen | A4 Markdown scaffold | Tier B | 0 | 4.7 | claude-code | | Governance-first AI coding framework with classification-locked phase gates, SSoT state machine, and 26 commands + 14 skills that enforce No Evidence = No Completion. |
| CCCP / gendosu-claude-plugins cccp-gendosu | A4 Markdown scaffold | Tier B | 0 | 0.5 | claude-code | | Provide a minimal plugin marketplace with TDD/git workflow commands and a meta-skill for creating new skills. |
| trilogy-group/codex-second-brain-starter-kit codex-second-brain-starter | A10 Memory engine | Tier B | 0 | 4.8 | openai-codex | UI | Enterprise Obsidian-based second brain builder for OpenAI Codex using BASB methodology, gpt-5.5 with reasoning-effort routing, and Tyler provider gateway for durable execution. |
| gg (2-letter agent-native git client) gg-cyclecore | A19 Token-efficient encoding | Tier B | 0 | 2 | unknown | CLI | Reduce AI agent token consumption by 96–98% on git/npm/brew operations by acting as a deterministic command broker that returns structured MCP payloads instead of full package schemas. |
| reqtext reqtext | A8 Cross-runtime harness | Tier B | 0 | 0.5 | unknown | CLI UI | Git-native requirements management CLI generating dual human+AI documentation from a flat JSON hierarchy with built-in test traceability. |
| Specorator (Luis85/agentic-workflow) specorator-workflow | A4 Markdown scaffold | Tier B | 0 | 6.8 | claude-code | CLI UI MCP | 18 commands + 40 skills + 36 specialist agents + 11-stage lifecycle covering Idea through Retrospective — full software product lifecycle with dedicated orchestrator, ADR system, GitHub MCP, and Obsidian companion for spec authoring. |
| trw-mcp (TRW Framework) trw-mcp | A3 MCP-anchored | Tier B | 0 | 9.1 | claude-code | CLI MCP | Prevent AI coding session state loss and quality drift by enforcing ceremony, capturing learnings, and orchestrating multi-agent teams through a 6-phase agile model. |
| Vibe Compiler (vibec) vibe-compiler | A6 Multi-agent orchestrator | Tier B | 0 | 0 | unknown | | |
| vitepress-plugin-openspec vitepress-plugin-openspec | A4 Markdown scaffold | Tier B | 0 | 0 | vitepress | | Render an OpenSpec directory as structured VitePress documentation pages with auto-generated sidebar and nav. |
| oh-my-claude-code (zephyrpersonal) zephyrpersonal-oh-my-claude | A6 Multi-agent orchestrator | Tier B | 0 | 1 | claude-code | | Cost-aware multi-agent orchestration for Claude Code with magic-keyword Ultrawork mode and YAML-metadata-driven agent routing. |
| AdaL adal | A6 Multi-agent orchestrator | Tier B | — | 0 | unknown | UI | Multi-agent engineering system with specialized Worker Agents (coding, research, browser use, review) and an Engineer Agent coordinator for teams |
| Aegis (Folder Trees) aegis-folder-trees | A4 Markdown scaffold | Tier B | — | 0 | unknown | | Community-referenced methodology using hierarchical folder trees as AI context and spec delivery mechanism. |
| Auto Claude auto-claude | A13 Methodology | Tier B | — | 0 | unknown | | unknown — Reddit reference only, no public repository found |
| Axon axon | A4 Markdown scaffold | Tier B | — | 0 | unknown | | unknown — repository not accessible |
| Blackbox Code blackbox-code | A4 Markdown scaffold | Tier B | — | 0 | unknown | | unknown — repository not found (HTTP 404) |
| Boomerang Tasks boomerang-tasks | A13 Methodology | Tier B | — | 0 | unknown | | Orchestrator-subagent dispatch pattern for task breakdown — not a standalone framework. |
| Claude Oracle claude-oracle | A4 Markdown scaffold | Tier B | — | 0 | unknown | | unknown — repository not accessible |
| claude-plugin-sdd (joestump) claude-plugin-design-joestump | A4 Markdown scaffold | Tier B | — | 0 | unknown | | |
| Design OS design-os | A13 Methodology | Tier B | — | 0 | unknown | | Onboarding AI agents and developers into large organically-grown ('vibe-coded') monolith codebases. |
| Desktop Commander desktop-commander | A3 MCP-anchored | Tier B | — | 0 | unknown | | Iterative PRD update tool referenced in Reddit — likely a usage pattern of Desktop Commander MCP server for file-based PRD management, not a dedicated PM framework. |
| Flow Framework flow-framework-khgs | A4 Markdown scaffold | Tier B | — | 0 | unknown | | unknown — repository not accessible |
| Forge (sudo-rebase) forge-sudo-rebase | A4 Markdown scaffold | Tier B | — | 0 | unknown | | unknown — repository not accessible |
| Knowledge Flywheel knowledge-flywheel | A13 Methodology | Tier B | — | 0 | unknown | | Session-compounding learnings extractor: each AI coding session feeds learnings back into future sessions, creating a compounding knowledge flywheel. |
| OpenClaw openclaw-os | A4 Markdown scaffold | Tier B | — | 0 | openclaw | | AI agent runtime platform supporting plugin-based tool installation (osmoda-bridge), --local execution, multi-provider model routing (provider/model format), and session management. |
| prd-to-code gist rbstp-prd-to-code | A13 Methodology | Tier B | — | 0.5 | unknown | | Three markdown files (Create PRD, Generate Tasks, Iterate Task) enforce the discipline of spec-before-code and atomic task execution. |
| Slate V1 slate-v1 | A4 Markdown scaffold | Tier B | — | 0 | unknown | | unknown — no public material available |
| Tars (Gemini CLI ecosystem) tars-gemini | A6 Multi-agent orchestrator | Tier B | — | 0 | gemini-cli | | Unknown — described as a local-first autonomous supervisor sidekick for the Gemini CLI ecosystem. |
| TaskVal CLI / Taskify Skill taskval | A13 Methodology | Tier B | — | 0 | unknown | | Validates task specifications before AI agent execution, refusing to proceed on vague or underspecified tasks — not a standalone locatable framework. |
| Vibe CEO'ing vibe-ceo-ing | A13 Methodology | Tier B | — | 0 | gemini | | A framing of BMAD-METHOD's web-UI planning phase for non-technical founders who want to use AI agents to produce project blueprints without writing code. |
| Ruflo (ruvnet) ruflo | A6 Multi-agent orchestrator | Unranked | 55,240 | 0 | unknown | | |
| Kiro kiro · ides | A5 Closed IDE | Seed | 3,762 | 2 | kiro | | Prevents 'vibe coding' by enforcing a Requirements→Design→Tasks spec pipeline with EARS notation before any code is generated. |
| ECC (Everything Claude Code) ecc-affaan · aggregation-packs | A2 Mirror cmd+skill | Tier A | 193,231 | 9.1 | claude-code | CLI UI | Comprehensive harness-native operator system: 246 skills + 61 agents + continuous learning hooks + multi-model routing across 8 AI tools. |
| Hermes Agent (NousResearch) hermes-agent · agent-sdks | A6 Multi-agent orchestrator | Tier A | 167,908 | 7.6 | standalone | CLI UI MCP | Self-improving personal AI agent with closed learning loop, 7 terminal backends, and messaging gateway — not tied to any AI coding tool. |
| OpenCode opencode · coding-agents | A6 Multi-agent orchestrator | Tier A | 165,457 | 5 | opencode-cli | CLI UI | Terminal-first AI coding agent with multi-model routing, native desktop app, and a typed .opencode/ configuration system for agents, commands, skills, and plugins. |
| Anthropic Skills (Official) anthropics-skills · skill-catalogs | A1 Skills-only | Tier A | 140,976 | 0 | claude-code | | Official Anthropic reference gallery for the Agent Skills standard, spanning creative, technical, and enterprise capability demonstrators. |
| Gemini CLI (Google) gemini-cli · coding-agents | A2 Mirror cmd+skill | Tier A | 104,604 | 2 | gemini-cli | CLI UI | Bring the full power of Gemini into the terminal with a free tier, Google Search grounding, and extensible MCP support. |
| OpenAI Codex CLI codex-cli · coding-agents | A2 Mirror cmd+skill | Tier A | 85,783 | 3.7 | codex-cli | CLI UI | Give developers a sandboxed, locally-running OpenAI coding agent with approval gates and skill orchestration. |
| Zed zed · ides | A5 Closed IDE | Tier A | 83,799 | 5 | zed | CLI UI | Open-source, GPU-accelerated IDE with a native AI agent featuring LSP-aware tools, update_plan for visible task tracking, spawn_agent for subagent delegation, and SKILL.md for reusable instructions. |
| claude-mem (thedotmack) claude-mem · memory-tools | A8 Cross-runtime harness | Tier A | 78,248 | 4 | claude-code | CLI UI MCP | Background worker service captures every tool call as an observation, AI-compresses sessions, and auto-injects relevant past context using a 3-layer progressive disclosure protocol — no re-explaining the codebase. |
| OpenHands openhands · coding-agents | A6 Multi-agent orchestrator | Tier A | 74,904 | 4.9 | openhands-gui | CLI UI | Open-source AI software development platform (open-source Devin alternative) with Docker sandbox isolation, 77.6% SWE-bench score, and 25+ trigger-activated domain skills. |
| Daytona daytona · sandboxes | A9 Sandbox substrate | Tier A | 72,455 | 4.5 | unknown | CLI UI MCP | Provide secure, elastic, sub-90ms sandbox compute infrastructure for running AI-generated code, accessible via multi-language SDKs and REST API. |
| Cline cline · coding-agents | A5 Closed IDE | Tier A | 62,338 | 4.3 | vscode | CLI UI | SDK-first autonomous coding agent with human-in-the-loop approval across VS Code, JetBrains, CLI, and a parallel-agent Kanban board. |
| gpt-engineer gpt-engineer · coding-agents | A6 Multi-agent orchestrator | Tier A | 55,211 | 0.9 | openai-api | CLI | ARCHIVED — the original LLM-driven code generation experiment that generates entire applications from a prose specification file. |
| pi (badlogic/earendil) pi-coding-agent · coding-agents | A8 Cross-runtime harness | Tier A | 55,166 | 4.7 | pi-cli | CLI UI | A minimal, hackable, multi-provider terminal coding agent that adapts to your workflows via npm-installable TypeScript Extensions and Markdown Skills — without forking. |
| MemPalace mempalace · memory-tools | A10 Memory engine | Tier A | 52,851 | 3 | claude-code | CLI MCP | Verbatim local-first AI memory with 96.6% R@5 retrieval on LongMemEval using zero API calls — structured into a palace hierarchy (wings/rooms/drawers) with a temporal knowledge graph backed by SQLite. |
| Goose (Block/AAIF) goose · coding-agents | A12 UI passthrough | Tier A | 45,882 | 4.5 | goose-cli | CLI UI MCP | General-purpose AI agent (not just code) with security-first tool inspection, recipe-based shareable configurations, and 15+ LLM provider support. |
| Agent Skills (Addy Osmani) agent-skills-addyosmani · skill-catalogs | A8 Cross-runtime harness | Tier A | 45,847 | 2.5 | claude-code | | Encodes senior-engineer software development lifecycle as 23 auto-routed skills and 7 slash commands for any AI coding agent. |
| Aider aider · coding-agents | A6 Multi-agent orchestrator | Tier A | 45,340 | 4.6 | anthropic-api | CLI UI | Terminal-native AI pair programming with git-first edit semantics and automatic commit messages. |
| wshobson/agents Plugin Marketplace wshobson-agents · skill-catalogs | A8 Cross-runtime harness | Tier A | 35,958 | 4.6 | claude-code | | Single Markdown source for 83 domain-specialized plugins that auto-generates idiomatic artifacts for five AI coding harnesses. |
| TabbyML/Tabby tabby · coding-agents | A8 Cross-runtime harness | Tier A | 33,542 | 3 | vscode | CLI UI | Self-hosted AI coding assistant server (alternative to GitHub Copilot) with admin dashboard, RAG-based completions, and multi-IDE support. |
| Continue continue · coding-agents | A6 Multi-agent orchestrator | Tier A | 33,391 | 1.4 | github-actions | CLI UI | AI code review as source-controlled, enforceable CI checks — green/red GitHub status checks powered by markdown-defined agent instructions. |
| Vibe Kanban vibe-kanban · ui-dashboards | A12 UI passthrough | Tier A | 26,518 | 4.3 | claude-code | CLI UI | Eliminate the overhead of planning, switching between agent terminals, and reviewing diffs by providing a single web dashboard with kanban, workspaces, diff review, and PR creation. |
| Beads (Yegge) beads-yegge · memory-tools | A10 Memory engine | Tier A | 24,094 | 5.9 | claude-code | CLI MCP | Dolt-powered distributed graph issue tracker where AI agents track tasks with hierarchical IDs and dependency edges, claim work atomically, store persistent memories via bd remember, and compact old issues via graceful decay summarization. |
| deepagents (LangChain) deepagents-langchain · agent-sdks | A10 Memory engine | Tier A | 23,351 | 6.5 | LangGraph | CLI | Opinionated Python agent harness on top of LangGraph with sub-agents, filesystem, memory, and context compaction bundled in |
| VoltAgent awesome-claude-code-subagents voltagent-subagents · skill-catalogs | A6 Multi-agent orchestrator | Tier A | 20,554 | 1.6 | claude-code | | 154+ domain-specialized Claude Code subagents across 10 categories, installable as plugins with no commands, skills, or hooks — pure agent catalog. |
| OpenAI Skills (Official) openai-skills · skill-catalogs | A1 Skills-only | Tier A | 20,338 | 1 | codex | | Tiered library of integration and workflow skills for Codex, with a self-bootstrapping skill-installer that lets agents discover and expand their own capabilities in-session. |
| SWE-agent swe-agent · coding-agents | A6 Multi-agent orchestrator | Tier A | 19,314 | 5.4 | anthropic-api | CLI | Research-grade GitHub issue fixing agent with YAML-configurable prompts, one-command-at-a-time REPL loop, and batch evaluation support for SWE-bench benchmarking. |
| agentmemory agentmemory · memory-tools | A10 Memory engine | Tier A | 18,016 | 3.8 | claude-code | CLI UI MCP | Persistent, searchable memory for AI coding agents that captures every tool interaction, compresses it via LLM, and injects relevant context into future sessions — so agents remember everything across session boundaries. |
| CUA cua-sandbox · sandboxes | A9 Sandbox substrate | Tier A | 17,104 | 3.2 | Claude Code | CLI UI MCP | Unified SDK for building, benchmarking, and deploying agents that interact with full OS GUIs via isolated VMs. |
| alirezarezvani/claude-skills alirezarezvani-claude-skills · skill-catalogs | A18 Self-evolving | Tier A | 16,211 | 2.5 | claude-code | MCP | 313+ skills for 12 AI tools covering engineering, marketing, C-level advisory, compliance, research, and finance — all from one canonical skill library. |
| Anthropic Knowledge Work Plugins anthropics-knowledge-work-plugins · skill-catalogs | A4 Markdown scaffold | Tier A | 16,165 | 1.6 | claude-cowork | MCP | Role-specialized plugin bundles with live MCP connectors that turn Claude into a domain expert for enterprise knowledge workers. |
| OpenHarness openharness · agent-sdks | A11 Governance | Tier A | 13,115 | 4.3 | standalone | CLI UI | Open-source Python agent runtime providing complete harness infrastructure: tools, memory, governance, swarm coordination, and messaging gateway. |
| E2B e2b · sandboxes | A9 Sandbox substrate | Tier A | 12,361 | 1 | unknown | CLI | Run AI-generated code safely in cloud-hosted isolated sandboxes via a 3-line SDK integration. |
| Trae Agent trae-agent · ides | A11 Governance | Tier A | 11,591 | 6.1 | trae-cli | CLI | Research-friendly open-source CLI coding agent by ByteDance, designed for academic ablation studies and modular LLM provider swapping. |
| Qodo (PR-Agent) qodo · ides | A8 Cross-runtime harness | Tier A | 11,345 | 2.4 | github | CLI | Open-source AI PR reviewer with single-call tool architecture, PR compression for large diffs, self-reflection quality gate, and cross-platform git provider support. |
| Superset superset · ui-dashboards | A8 Cross-runtime harness | Tier A | 11,262 | 3.5 | claude-code | UI MCP | macOS desktop app that eliminates context-switching overhead when running 10–100+ parallel AI coding agents, each isolated in its own git worktree, monitored from one kanban dashboard. |
| OpenSandbox opensandbox · sandboxes | A9 Sandbox substrate | Tier A | 10,828 | 3 | Claude Code | CLI MCP | Protocol-first general-purpose sandbox platform for AI applications with multi-language SDKs and pluggable isolation backends. |
| Context-Engineering Handbook context-engineering-handbook · methodology-docs | A13 Methodology | Tier A | 9,009 | 0 | claude-code | | Provides a first-principles, research-grounded vocabulary and learning path for context engineering — the discipline of designing the complete information payload an LLM receives, beyond single prompts. |
| Sweep AI sweep · coding-agents | A11 Governance | Tier A | 7,712 | 2.6 | GitHub | CLI | Autonomous GitHub bot that converts issues to pull requests using a sequential multi-agent pipeline. |
| walkinglabs/learn-harness-engineering learn-harness-engineering-walkinglabs · methodology-docs | A13 Methodology | Tier A | 6,558 | 0.9 | generic | UI | Teach harness engineering from first principles (12 lectures + 6 projects) and provide a scaffolding skill (harness-creator) that creates the five subsystems any reliable AI coding agent needs. |
| Microsandbox microsandbox · sandboxes | A9 Sandbox substrate | Tier A | 6,299 | 1 | unknown | CLI MCP | Spawn hardware-isolated microVMs as child processes directly from application code, with no server setup, in under 100ms. |
| CubeSandbox cubesandbox · sandboxes | A9 Sandbox substrate | Tier A | 5,940 | 1 | Claude Code | CLI UI | Sub-60ms KVM microVM sandboxes for AI agents with E2B drop-in compatibility and <5MB memory overhead. |
| 1Code 1code · ui-dashboards | A12 UI passthrough | Tier A | 5,542 | 5.5 | claude-code | UI MCP | Cursor-like desktop experience for Claude Code and Codex with cloud background agents, event-driven automations, and a full IDE-quality git client. |
| worktrunk worktrunk · sandboxes | A8 Cross-runtime harness | Tier A | 5,197 | 2.4 | claude-code | CLI UI | Make git worktree lifecycle (create, switch, list, merge, cleanup) as simple as branch operations, designed for managing 5-10+ parallel AI agents. |
| sandcastle (mattpocock) sandcastle-mattpocock · agent-sdks | A9 Sandbox substrate | Tier A | 5,103 | 3 | claude-code | CLI | Container-isolated TypeScript SDK for orchestrating AI coding agents with Docker/Podman/Vercel Firecracker sandboxes and git-worktree branch management. |
| mirrord mirrord · linters | A8 Cross-runtime harness | Tier A | 5,089 | 2.2 | none (multi-tool, infrastructure layer) | CLI UI | Routes local process syscalls (network, files, DNS, env) through a live Kubernetes cluster pod so AI agents test against real infrastructure without deploying. |
| VoltAgent awesome-codex-subagents voltagent-codex-subagents · skill-catalogs | A9 Sandbox substrate | Tier A | 4,894 | 2.6 | codex-cli | | 166 Codex-native TOML subagents across 13 categories with sandbox_mode isolation and smart GPT model routing. |
| mini-swe-agent mini-swe-agent · coding-agents | A8 Cross-runtime harness | Tier A | 4,526 | 3.2 | mini-swe-agent-cli | CLI UI | Prove that a ~130-line agent with only a bash tool can achieve >74% on SWE-bench, and serve as a clean research baseline + hackable daily tool. |
| Micro Agent (Builder.io) micro-agent · coding-agents | A8 Cross-runtime harness | Tier A | 4,307 | 2.4 | micro-agent-cli | CLI | Generate code that passes tests, nothing else — the smallest possible TDD agent that avoids the Roomba-under-the-table problem of general coding agents. |
| Mistral Vibe mistral-vibe · ides | A8 Cross-runtime harness | Tier A | 4,284 | 0.5 | vibe | CLI UI | Mistral AI's first-party open-source CLI coding agent with configurable safety profiles, subagent delegation, SKILL.md custom skills, and ACP programmatic backend support. |
| Flue flue · agent-sdks | A8 Cross-runtime harness | Tier A | 3,731 | 3 | Node.js | CLI | TypeScript server framework for building deployable headless agent services, with first-class Cloudflare Workers + Durable Objects support |
| Basic Memory basic-memory · memory-tools | A10 Memory engine | Tier A | 3,087 | 0.5 | claude-code | CLI UI MCP | Gives AI agents a persistent, human-readable knowledge graph of project decisions, observations, and relations stored as plain Markdown files. |
| Crystal (stravu) crystal · ui-dashboards | A12 UI passthrough | Tier A | 3,068 | 2.8 | claude-code | UI | Manage multiple parallel AI coding sessions in isolated git worktrees from a single desktop GUI. |
| Maestro (RunMaestro) maestro-runmaestro · ui-dashboards | A12 UI passthrough | Tier A | 2,954 | 6 | claude-code | CLI UI | Orchestrate unlimited parallel AI agent sessions with a keyboard-first desktop app including Group Chat coordination and Auto Run playbooks. |
| Awesome Harness Engineering (walkinglabs) awesome-harness-engineering-walkinglabs · methodology-docs | A13 Methodology | Tier A | 2,734 | 0 | generic | | Curate the authoritative reference list of articles, benchmarks, and tools for harness engineering — the practice of shaping the environment around AI agents so they work reliably. |
| Claude-Supermemory claude-supermemory · memory-tools | A10 Memory engine | Tier A | 2,584 | 0.5 | claude-code | | Gives Claude Code team-level persistent memory via a cloud API, sharing project knowledge across developers and sessions without local infrastructure. |
| wshobson/commands Slash Command Collection wshobson-commands · skill-catalogs | A8 Cross-runtime harness | Tier A | 2,482 | 2.1 | claude-code | | 57 slash commands (15 multi-agent workflows + 42 single-purpose tools) for Claude Code, now superseded by the wshobson/agents plugin marketplace. |
| Agent Sandbox (kubernetes-sigs) agent-sandbox-k8s · sandboxes | A9 Sandbox substrate | Tier A | 2,453 | 3.5 | unknown | | Provide a first-class Kubernetes CRD for singleton stateful workloads with stable identity, persistent storage, and hibernation — the missing primitive for AI agent runtimes. |
| Agent Governance Toolkit (microsoft) agent-governance-toolkit · linters | A11 Governance | Tier A | 2,309 | 4.4 | claude-code | CLI UI MCP | Enterprise-grade AI agent governance: YAML policy enforcement, 12-vector prompt injection defense, zero-trust identity, tamper-evident audit logging, and OWASP Agentic Top 10 coverage for production AI agents. |
| jeremylongshore/claude-code-plugins-plus-skills (Tons of Skills) jeremylongshore-plugins-skills · skill-catalogs | A6 Multi-agent orchestrator | Tier A | 2,242 | 3.2 | claude-code | CLI UI MCP | 431 plugins, 2,810 skills, 200 agents with a dedicated npm CLI (ccpi) and marketplace website — the Claude Code plugin ecosystem treated as a commercial software product. |
| RA.Aid ra-aid · coding-agents | A10 Memory engine | Tier A | 2,224 | 5.2 | ra-aid-cli | CLI UI | Autonomous software development via three specialized agents (Research → Planning → Implementation) with SQLite memory and on-demand expert-model escalation. |
| lean-ctx lean-ctx · memory-tools | A19 Token-efficient encoding | Tier A | 2,186 | 4.6 | cursor | CLI UI MCP | A full-session context runtime that compresses file reads (10 modes), shell output (60+ patterns), and session memory (CCP) to save 60-99% of tokens while providing real-time observability. |
| TDD Guard tdd-guard · linters | A11 Governance | Tier A | 2,149 | 3.9 | claude-code | CLI | Mechanically enforces the Red-Green-Refactor TDD cycle by blocking file writes that violate TDD principles via a PreToolUse hook and LLM semantic validator. |
| AgentsMesh agentsmesh · ui-dashboards | A12 UI passthrough | Tier A | 2,137 | 5.5 | claude-code | CLI UI MCP | Multi-tenant workforce platform that gives every team member a squad of AI coding agents coordinated through channels, pod bindings, and a kanban task board, with on-prem runners keeping code on your infrastructure. |
| MCP Shrimp Task Manager mcp-shrimp-task-manager · memory-tools | A3 MCP-anchored | Tier A | 2,103 | 2.7 | cursor | UI MCP | Convert natural language requests into structured AI development tasks with chain-of-thought enforcement, reflection gates, and dependency tracking. |
| memsearch CC Plugin memsearch-cc · memory-tools | A10 Memory engine | Tier A | 1,841 | 1.3 | claude-code | CLI | Give AI coding agents automatic cross-session semantic memory backed by local Markdown files and a rebuildable Milvus vector index — no API key or external database required. |
| TmuxAI tmuxai · coding-agents | A8 Cross-runtime harness | Tier A | 1,833 | 4.2 | tmuxai-cli | CLI UI | Be the AI pair programmer that watches your tmux screen and helps with any terminal workflow — no shell wrappers, no workflow changes, just observe and assist. |
| VibeKit vibekit · sandboxes | A9 Sandbox substrate | Tier A | 1,790 | 4.2 | claude-code | CLI | TypeScript SDK that routes any coding agent to any sandbox backend with built-in secrets redaction, git worktree isolation, and PR creation. |
| Open Cowork open-cowork · ui-dashboards | A12 UI passthrough | Tier A | 1,417 | 2.5 | claude-code | UI | Makes Claude Code accessible to knowledge workers on Windows/macOS with one-click install, VM sandbox, document skills, and IM remote control. |
| deepagentsjs deepagentsjs · agent-sdks | A8 Cross-runtime harness | Tier A | 1,262 | 6.5 | LangGraph.js | | TypeScript port of LangChain's Python Deep Agents harness, with advanced generic type safety for agent configuration |
| Maestro (its-maestro-baby) maestro-its-maestro-baby · ui-dashboards | A12 UI passthrough | Tier A | 1,152 | 3.3 | claude-code | CLI UI MCP | Run 1-6 AI coding sessions in isolated git worktrees with native Tauri performance and a reverse-MCP status protocol for agent state reporting. |
| daymade/claude-code-skills daymade-cc-skills · skill-catalogs | A1 Skills-only | Tier A | 1,111 | 2.1 | claude-code | | 52 production-hardened skills with emphasis on skill craftsmanship — a fork of Anthropic's official skill-creator with security gates, structured UX checkpoints, and quantitative quality benchmarks. |
| IWE iwe · memory-tools | A10 Memory engine | Tier A | 1,086 | 0.5 | claude-code | CLI MCP | Turns a directory of Markdown files into a queryable knowledge graph with LSP for editors and MCP for AI agents — no database, no lock-in, git-versionable. |
| Clearwing clearwing · linters | A9 Sandbox substrate | Tier A | 982 | 7 | none | CLI UI MCP | Autonomous LLM-powered vulnerability discovery with crash-validated PoCs, multi-stage model routing, and responsible disclosure tooling — open-source reimplementation of Anthropic's Glasswing. |
| mini-coding-agent mini-coding-agent · coding-agents | A4 Markdown scaffold | Tier A | 882 | 3.7 | ollama | CLI | A single-file zero-dependency Python coding agent that demonstrates the six core components of coding agents for educational purposes, running entirely on local Ollama models. |
| Arrakis arrakis · sandboxes | A9 Sandbox substrate | Tier A | 816 | 3.3 | unknown | CLI UI MCP | Self-hosted MicroVM sandbox server with snapshot-and-restore for AI agent code execution, computer use (VNC/Chrome), and MCTS-style backtracking. |
| context-space context-space · memory-tools | A8 Cross-runtime harness | Tier A | 810 | 1.4 | claude-code | MCP | Provides production-grade OAuth-secured MCP tool aggregation across 14+ external services, eliminating the hardest part of connecting AI agents to real-world APIs. |
| Codexia codexia · ui-dashboards | A16 Cross-vendor router | Tier A | 690 | 5.7 | openai-codex | UI | Tauri desktop app providing visual control plane, task scheduler, git worktree manager, and headless REST API for Codex CLI + Claude Code sessions. |
| kiro (jasonkneen) kiro-jasonkneen · ides | A8 Cross-runtime harness | Tier A | 665 | 1.5 | claude-code | MCP | Verbatim Kiro IDE system prompt + 8 spec-driven development skills packaged as a Claude Code plugin for teams that want Kiro's methodology without the paid IDE. |
| Parallel Code parallel-code · ui-dashboards | A12 UI passthrough | Tier A | 650 | 4.9 | claude-code | UI | Dispatch ten parallel AI coding agents in isolated git worktrees (or Docker containers) with AI Arena head-to-head comparison and per-task coverage tracking. |
| beads-ui beads-ui · memory-tools | A4 Markdown scaffold | Tier A | 621 | 0 | beads-bd | CLI UI | Give human developers a browser-based kanban view of Beads issues while AI agents work in the CLI. |
| leash (strongdm) leash-strongdm · linters | A11 Governance | Tier A | 565 | 3.9 | claude-code | CLI UI | Wraps AI coding agents in containers with eBPF-enforced Cedar policies, making policy violations (unauthorized file access, network connections, MCP tool calls) structurally impossible rather than probabilistically unlikely. |
| GroundZero Package Manager (OpenPackage / opkg) gpm-groundzero · standalone-clis | A8 Cross-runtime harness | Tier A | 557 | 0.9 | claude-code | CLI UI | Universal CLI package manager for AI coding agent configuration files (rules, commands, agents, skills, MCPs) with cross-platform conversion for 40+ tools. |
| TheAuditor theauditor · linters | A4 Markdown scaffold | Tier A | 550 | 0 | unknown | | AI context and enforcement layer for code-generating agents (pre-launch; no public source). |
| SWE-ReX swe-rex · sandboxes | A8 Cross-runtime harness | Tier A | 508 | 0 | SWE-agent | | Unified Python library for interacting with persistent bash sessions in any execution backend (local, Docker, Fargate, Modal). |
| Sponsio sponsio · linters | A11 Governance | Tier A | 445 | 3.7 | claude-code | CLI UI | Deterministic runtime enforcement library checking every AI agent tool call against pure-code contracts at p50 0.139ms with zero LLM cost. |
| Open Agent (Th0rgal) open-agent-thorgal · sandboxes | A8 Cross-runtime harness | Tier A | 438 | 5 | Claude Code | UI | Same as sandboxed-sh — self-hosted AI agent orchestrator (non-canonical slug). |
| sandboxed.sh sandboxed-sh · sandboxes | A8 Cross-runtime harness | Tier A | 438 | 5 | Claude Code | UI | Self-hosted orchestrator for AI coding agents with isolated workspaces, multi-runtime support, and Library-based configuration management. |
| Vet (Verify Everything) vet-imbue · linters | A8 Cross-runtime harness | Tier A | 385 | 0.4 | claude-code | CLI | Catches intent-implementation mismatches in AI-generated code by cross-examining git diffs against the stated goal and conversation history. |
| pydantic-ai-harness pydantic-ai-harness · agent-sdks | A99 Unclassified | Tier A | 354 | 0 | Pydantic AI | | Official capability extension library for Pydantic AI agents — bundles tools, hooks, and lifecycle primitives into pip-installable capabilities, with CodeMode collapsing N tool calls into 1 sandboxed Python execution |
| crit crit-review · linters | A8 Cross-runtime harness | Tier A | 350 | 4.5 | claude-code | CLI UI | Brings GitHub PR-style human inline review to AI agent output (plans, code, live apps, HTML) with persistent per-line comments and round-to-round diffs, without a GitHub round-trip. |
| Dorothy dorothy · ui-dashboards | A12 UI passthrough | Tier A | 291 | 4.5 | claude-code | UI MCP | Local-first autonomous agent platform with Super Agent MCP orchestration, memory REST API, GitHub/JIRA automations with deduplication, and Telegram/Slack remote control. |
| Water water · agent-sdks | A8 Cross-runtime harness | Tier A | 288 | 1 | any (framework-agnostic) | CLI | Python agent harness providing infrastructure (orchestration, resilience, observability, fallback chains) around any AI agent framework |
| Capsule capsule · sandboxes | A9 Sandbox substrate | Tier A | 285 | 3 | none | CLI | WebAssembly-based function-level sandbox for executing untrusted Python/TypeScript code with configurable CPU, memory, timeout, and network limits. |
| OmniCoreAgent omnicore-agent · agent-sdks | A8 Cross-runtime harness | Tier A | 241 | 4 | LiteLLM (provider-agnostic) | CLI | Python production agent harness with parallel tool batching, structured observations, BM25 tool retrieval, signature loop detection, and modular production backends |
| claude-self-reflect claude-self-reflect · memory-tools | A10 Memory engine | Tier A | 214 | 1 | claude-code | CLI MCP | Gives Claude Code persistent conversation memory via a zero-dependency 44MB Rust binary with 6 lifecycle hooks and a 3-layer progressive enrichment pipeline. |
| SymDex symdex · memory-tools | A10 Memory engine | Tier A | 190 | 0.5 | claude-code | CLI MCP | Turns a repository into a local symbolic index of symbols, routes, and call graphs so AI agents retrieve exact code instead of scanning whole files. |
| Antigravity Link antigravity-link · ides | A12 UI passthrough | Tier A | 180 | 0 | antigravity | UI MCP | Mobile companion and MCP/HTTP bridge for Google's Antigravity IDE, enabling remote monitoring, control, and agent automation via Chrome DevTools Protocol. |
| nexu-io/harness-engineering-guide nexu-harness-guide · methodology-docs | A13 Methodology | Tier A | 134 | 0 | generic | UI | Provide a practical, code-first reference guide to harness engineering — from first principles to production patterns — maintained by a managed agent platform company. |
| Nemp Memory nemp-memory · memory-tools | A19 Token-efficient encoding | Tier A | 101 | 1.7 | claude-code | | Persists AI agent context across sessions as 100%-local plain JSON files with zero dependencies, zero cloud, and agent identity attribution. |
| Kagan kagan · ui-dashboards | A16 Cross-vendor router | Tier A | 88 | 5.9 | claude-code | CLI UI MCP | Kanban TUI for AI coding agents with a structurally enforced human review gate (REVIEW → DONE cannot be automated) — one git worktree per task, 14 agent backends, Textual TUI + React web + VS Code extension all sharing SQLite state. |
| Claude Code Guardrails claude-code-guardrails · linters | A11 Governance | Tier A | 54 | 3 | claude-code | | Automates git safety in Claude Code sessions: blocks writes to protected branches, snapshots every change as a micro-commit, and squashes session checkpoints into clean task commits. |
| VibeSec vibesec · linters | A4 Markdown scaffold | Tier A | 46 | 0.5 | cursor | | Injects OWASP-aligned security rules into Cursor and Windsurf AI assistants to prevent common security vulnerabilities in AI-generated code. |
| ovachiever/droid-tings droid-tings · skill-catalogs | A6 Multi-agent orchestrator | Tier A | 44 | 2.1 | claude-code | | 375 aggregated skills + 155 droids from 15+ repos covering AI/ML, scientific research, and development — targeting both Claude Code and Factory's Droid system. |
| Nex nex-as-skill · memory-tools | A10 Memory engine | Tier A | 44 | 1.3 | claude-code | CLI MCP | Turns AI agent conversations into a cloud-backed entity knowledge graph accessible from any AI platform, with automatic context injection before every prompt and automatic capture after every session. |
| knowhub knowhub · standalone-clis | A13 Methodology | Tier A | 40 | 0 | generic | CLI | Synchronize AI coding-agent knowledge files (rules, guidelines, templates) from a central source to multiple AI-tool-specific destinations via copy or symlink. |
| spec-compare (cameronsjo) spec-compare · methodology-docs | A13 Methodology | Tier A | 39 | 0 | none | | Curated research comparison of 12 spec-driven development tools with graded use-case scoring, decision frameworks, and the SDD Maturity Levels taxonomy (Spec-First / Spec-Anchored / Spec-as-Source). |
| MemoryAgent memoryagent · memory-tools | A8 Cross-runtime harness | Tier A | 38 | 0.5 | claude-code | | Implements memory as a plain text file managed by the agent's native Read/Write/Edit/Grep tools, with a structured analyze command that produces a 7-section intelligence briefing. |
| Brood Box brood-box · sandboxes | A8 Cross-runtime harness | Tier A | 36 | 2.7 | claude-code | CLI | Wrap any coding agent in a hardware-isolated microVM with COW workspace snapshot, egress firewall, and interactive change review. |
| RLM (Infinite Memory) rlm-claude · memory-tools | A3 MCP-anchored | Tier A | 35 | 2.5 | claude-code | MCP | Prevents context loss from Claude Code's /compact by intercepting compaction events and providing a 14-tool MCP server for persistent session memory with hybrid search. |
| Kratos MCP kratos-mcp · memory-tools | A10 Memory engine | Tier A | 34 | 0.5 | claude-desktop | CLI MCP | Ultra-lean 12-tool MCP memory server for any AI coding tool with per-project SQLite isolation and built-in PII detection. |
| AgentLint agentlint · linters | A11 Governance | Tier A | 33 | 2.6 | claude-code | CLI | Scores agent harness files (CLAUDE.md, AGENTS.md, hooks, CI) against 58 evidence-backed checks to quantify harness quality and provide guided fix plans. |
| CogniLayer v4 cognilayer · memory-tools | A19 Token-efficient encoding | Tier A | 28 | 4.1 | claude-code | CLI UI MCP | Provides AI coding agents with typed semantic memory, tree-sitter code intelligence, and a multi-agent coordination protocol to save 80-200K+ tokens per session. |
| CCCMemory MCP (xiaolai) cccmemory-mcp · memory-tools | A10 Memory engine | Tier A | 26 | 3 | claude-code | CLI MCP | SQLite + hybrid vector+FTS5 MCP server providing 40+ memory tools (conversation indexing, decision tracking, working memory, session handoff, tag management) for cross-session AI agent memory. |
| AI-AfterImage ai-afterimage · memory-tools | A10 Memory engine | Tier A | 22 | 3.4 | claude-code | CLI | Episodic code memory that intercepts every Write/Edit, injects similar past code via a deny-then-allow pattern, and tracks file churn to warn before modifying stable code or repeatedly editing the same function. |
| Eval Marketplace eval-marketplace · linters | A4 Markdown scaffold | Tier A | 22 | 0.4 | claude-code | | Automated supply-chain security evaluation of third-party agent skills and MCP servers before installation, using GitHub and web-scraping MCP tools. |
| MartinLoop martinloop · linters | A11 Governance | Tier A | 22 | 6 | claude-code | CLI MCP | Governance control plane that wraps Claude Code/Codex loops with hard budget caps, 11-class failure taxonomy, Red-Blue adversarial probes, safety leash, and JSONL audit records — stops the 'Ralph Loop' failure mode. |
| AgentTier agenttier · sandboxes | A11 Governance | Tier A | 19 | 3.5 | Claude Code | CLI UI | Kubernetes-native control plane for isolated, persistent sandbox environments for AI agents. |
| CC Marketplace (setouchi-h) cc-marketplace-setouchi · skill-catalogs | A4 Markdown scaffold | Tier A | 16 | 0.9 | claude-code | UI | Automate the repetitive git, PR, and build tasks that surround every Claude Code session without adding workflow methodology. |
| aiignore-cli aiignore-cli · standalone-clis | A8 Cross-runtime harness | Tier A | 8 | 0 | none (multi-tool) | CLI | One command to generate correct, security-researched ignore configurations for all AI coding tools in a project, with documented CVEs and bypass vulnerabilities per tool. |
| ctxlint ctxlint · linters | A3 MCP-anchored | Tier A | 6 | 0 | claude-code | CLI MCP | Pre-session static linter that validates AI context files against the actual codebase, catching broken paths, stale commands, secrets, and contradictions. |
| Ozzeron prompt-pack ozzeron-prompt-pack · skill-catalogs | A8 Cross-runtime harness | Tier A | 6 | 0.4 | cursor | | Prevent the recurring AI-generated technical debt patterns (duplicate artifacts, convention drift, scope creep) that accumulate regardless of which AI tool is used. |
| Hawkeye hawkeye · linters | A3 MCP-anchored | Tier A | 5 | 5.7 | claude-code | CLI UI MCP | Full-session flight recorder for AI coding agents: records every action to SQLite, detects objective drift, enforces guardrails via PreToolUse hooks, and exposes 27 MCP tools for agent self-awareness. |
| danielrosehill/AI-Harnesses danielrosehill-ai-harnesses · methodology-docs | A13 Methodology | Tier A | 3 | 0 | generic | | Point-in-time taxonomy snapshot (April 2026) of projects self-describing as AI agent harnesses, with etymology notes on the 'harness' vs 'framework' vs 'platform' distinction. |
| spec-driven-kiro-project (lotharschulz) spec-driven-kiro-project · ides | A4 Markdown scaffold | Tier A | 3 | 0.5 | kiro | | Example project demonstrating Kiro IDE's spec-driven development approach applied to a real TypeScript web application. |
| che-incubator demo-spec-driven-development-with-ai ra-aid-che-incubator · coding-agents | A8 Cross-runtime harness | Tier A | 2 | 0 | ra-aid | | Zero-friction template for spec-driven development in cloud workspaces: write a spec, run one command, get working code. |
| vibelint vibelint · linters | A4 Markdown scaffold | Tier A | 2 | 0.5 | claude-code | CLI | Post-session linter that audits Claude Code session transcripts for quality, cost, and config issues. |
| Panaversity SDD Exercises panaversity-sdd-exercises · methodology-docs | A13 Methodology | Tier A | 1 | 0 | claude-code | | Teach developers the four-phase SDD workflow (Research → Specification → Refinement → Implementation) through real-world planning scenarios before writing any code. |
| BeadsViewer beadsviewer · memory-tools | A10 Memory engine | Tier A | 0 | 0 | none | UI | Local Next.js 14 kanban + tree view for Beads task graphs, reading all data via bd --json API routes with drag-and-drop status changes and inline task editing. |
| cc-audit cc-audit · linters | A11 Governance | Tier A | 0 | 0.4 | claude-code | | CI gate that lints CLAUDE.md/AGENTS.md against a 12-rule baseline, detecting missing rules and leaked secrets in pull requests. |
| cc-simple-memory (STRML) cc-simple-memory · memory-tools | A4 Markdown scaffold | Tier A | 0 | 4 | claude-code | CLI | Minimal shell + Node.js plugin with 3 lifecycle hooks (SessionStart/Stop/PreCompact) that extracts session learnings into per-project and global Markdown files using tiered AI models (Haiku/Sonnet/Opus) based on task complexity. |
| kevinten-ai/agent-harness-guide kevinten-harness-guide · methodology-docs | A13 Methodology | Tier A | 0 | 0 | generic | | Provide a Chinese-language knowledge graph and 4-phase learning path for systematically mastering AI agent harness engineering. |
| Nanocoder nanocoder · coding-agents | A4 Markdown scaffold | Tier A | 0 | 0 | unknown | | unknown — repository not publicly accessible |
| TokRepo tokrepo · standalone-clis | A8 Cross-runtime harness | Tier A | 0 | 1 | claude-code | CLI MCP | Open registry infrastructure for AI assets where agents can autonomously discover, install, and contribute reusable capabilities across sessions and platforms. |
| Zenable zenable · linters | A11 Governance | Tier A | 0 | 0 | unknown | | Commercial AI governance/compliance product (zenable.io) — no public source available. |
| AgentsKB agentskb · memory-tools | A10 Memory engine | Tier A | — | 0 | unknown | MCP | Hosted knowledge base platform for AI agents to query, update, and manage structured knowledge. |
| Antigravity (Google) antigravity · ides | A12 UI passthrough | Tier A | — | 0 | antigravity | UI | Google's AI IDE (VS Code fork with integrated AI coding agent). |
| backgrounder.dev backgrounder-dev · ui-dashboards | A99 Unclassified | Tier A | — | 0 | unknown | UI | Hosted background coding agent interface (closed SaaS — insufficient public material for full analysis). |
| Cline Memory Bank (official) cline-memory-bank-official · methodology-docs | A13 Methodology | Tier A | — | 0.5 | cline | | Official Cline documentation for the memory bank pattern: six hierarchical Markdown files providing cross-session project context via custom instructions. |
| Conductor.build conductor-build · ui-dashboards | A12 UI passthrough | Tier A | — | 3.5 | claude-code | UI | Native macOS desktop app that orchestrates multiple parallel Claude Code and Codex agents in isolated git worktrees with a visual monitoring and review UI. |
| ContextMCP contextmcp · memory-tools | A3 MCP-anchored | Tier A | — | 0 | unknown | MCP | Hosted MCP server aggregating tool access to external services for AI agents. |
| Cosine cosine · ides | A12 UI passthrough | Tier A | — | 1 | cosine-desktop | CLI UI | First-party AI coding platform with proprietary Lumen models trained for production-first coding (minimal diffs, niche languages, slop elimination), with desktop + CLI + cloud surfaces and air-gapped enterprise deployment. |
| Devin devin · ides | A12 UI passthrough | Tier A | — | 5.9 | devin | CLI UI MCP | Fully autonomous cloud AI engineer that handles complete tasks (tickets, migrations, PRs) in isolated microVMs without requiring the developer to be present. |
| Factory Droid droid-factory · ides | A8 Cross-runtime harness | Tier A | — | 3 | droid-cli | CLI UI | Full-platform AI software factory: hooks turn prompts into deterministic enforcement, Missions add structured orchestration, and Droid Computers provide persistent execution environments — targeting teams who want reproducible AI-assisted development, not ad-hoc prompting. |
| Entroly entroly · memory-tools | A10 Memory engine | Tier A | — | 3.6 | unknown | CLI UI | |
| FlyDex flydex · sandboxes | A4 Markdown scaffold | Tier A | — | 0 | unknown | | unknown — no public GitHub repository accessible |
| Frontman frontman · ides | A12 UI passthrough | Tier A | — | 0 | unknown | | unknown — insufficient public material |
| Headless Coder SDK headless-coder-sdk · agent-sdks | A4 Markdown scaffold | Tier A | — | 0 | unknown | | unknown — repository not found (HTTP 404) |
| Hermes IDE hermes-ide · coding-agents | A12 UI passthrough | Tier A | — | 0.4 | unknown | UI | |
| kiro spec agent (notdp gist) kiro-spec-agent-gist · ides | A13 Methodology | Tier A | — | 1.5 | claude-code | | Single CLAUDE.md file injecting the complete Kiro spec workflow (requirements → design → tasks → one-task-at-a-time execution) into Claude Code via system prompt. |
| maxim-saplin kiro-SDD gist maxim-saplin-kiro-sdd · ides | A13 Methodology | Tier A | — | 1.5 | cursor | | Dual-format (Cursor + GitHub Copilot) gist providing Kiro-style SDD slash commands without any installation requirements. |
| Stoneforge stoneforge · ui-dashboards | A6 Multi-agent orchestrator | Tier A | — | 4.7 | claude-code | CLI UI | Three-role (Director/Worker/Steward) orchestration platform with dependency-aware dispatch daemon, automatic merge-via-Steward, and a full web dashboard including persistent channels, document library, and metrics. |
| SWE-Pruner swe-pruner · memory-tools | A10 Memory engine | Tier A | — | 1.9 | unknown | CLI | |
| VibeBox vibebox · sandboxes | A4 Markdown scaffold | Tier A | — | 0 | unknown | | unknown — no public GitHub repository accessible |
| ECC — everything-claude-code everything-claude-code · aggregation-packs | A2 Mirror cmd+skill | Tier B | 194,119 | 6.3 | claude-code | CLI UI | Provide a complete harness-native operator system covering the full SDLC with cross-tool portability, continuous learning, and security hardening. |
| Anthropic Skills — IP Guard Delta Report anthropics-skills-ip-guard · skill-catalogs | A1 Skills-only | Tier B | 141,197 | 0 | claude-code | | No ip-guard skill found in the anthropics/skills repo; this entry is a delta report tracking its absence. |
| UI UX Pro Max ui-ux-pro-max · skill-catalogs | A1 Skills-only | Tier B | 83,052 | 0 | claude-code | CLI | Provides a searchable design database (67 styles, 161 palettes, 57 font pairings, 99 UX guidelines) with a reasoning engine that selects appropriate design choices based on product type. |
| LobeHub lobehub · ui-dashboards | A12 UI passthrough | Tier B | 77,741 | 7 | web-app | CLI UI | Full-stack multi-agent platform that acts as a Chief Agent Operator — scheduling, delegating, and reporting on an AI team running 7x24, with IM gateway delivery and white-box personal memory. |
| Paperclip paperclip · ui-dashboards | A12 UI passthrough | Tier B | 67,777 | 7.4 | own-server | CLI UI MCP | Corporate control plane for AI-agent companies — org charts, budgets, governance, heartbeat scheduling, and multi-company isolation so humans run a company of agents rather than managing individual agent sessions. |
| learn-claude-code learn-claude-code · methodology-docs | A13 Methodology | Tier B | 62,800 | 0 | claude-code | | 20-lesson trilingual harness engineering curriculum teaching Claude Code internals from agent loop fundamentals to autonomous teams, MCP plugins, and worktree isolation. |
| pi-mono pi-mono · coding-agents | A6 Multi-agent orchestrator | Tier B | 55,497 | 5 | pi-cli | CLI UI | Minimal, extensible TypeScript terminal coding agent with multi-provider LLM API, TUI library, and TypeScript Extension API — adapt to workflows without forking. |
| rtk (Real Token Killer) rtk-token-killer · standalone-clis | A19 Token-efficient encoding | Tier B | 54,590 | 2.1 | claude-code | CLI | Intercepts Claude Code's Bash tool calls at the PreToolUse hook and compresses verbose CLI output (git status, test runners, builds) by 60-90% before it enters Claude's context window, reducing token spend without changing the developer's workflow. |
| Agno agno · agent-sdks | A6 Multi-agent orchestrator | Tier B | 40,363 | 4.5 | any | UI | Build, deploy, and manage production agent platforms with full ownership of data, memory, and compute. |
| Multica multica · ui-dashboards | A8 Cross-runtime harness | Tier B | 33,320 | 5 | claude-code | CLI UI | Open-source managed agents platform with 4 UI surfaces (web/desktop/mobile/CLI), squad routing, autopilots, and skill compounding — turns coding agents into org-chart teammates supporting 11 different agent CLIs. |
| Claude Code Templates (davila7) claude-code-templates-davila · standalone-clis | A6 Multi-agent orchestrator | Tier B | 27,587 | 3.4 | claude-code | CLI UI | npm-installable marketplace of 600+ Claude Code agents, commands, hooks, MCPs, and settings for composable workflow assembly. |
| Claude Scientific Skills claude-scientific-skills · skill-catalogs | A11 Governance | Tier B | 26,062 | 0.6 | claude-code | | Provides 138 curated skills for scientific Python workflows across biology, chemistry, medicine, physics, and engineering — making AI coding agents significantly more reliable for domain-specific library usage. |
| Claudia GUI claudia-gui · ui-dashboards | A12 UI passthrough | Tier B | 21,948 | 0 | claude-code | UI | Visual desktop GUI for Claude Code session browsing, agent management, and cost tracking — identity resolved to opcode. |
| opcode opcode · ui-dashboards | A12 UI passthrough | Tier B | 21,948 | 1.2 | claude-code | UI | Visual desktop shell for Claude Code sessions, agents, usage, and checkpoints without any methodology overlay. |
| AGENTS.md agents-md · methodology-docs | A13 Methodology | Tier B | 21,710 | 0.5 | (all — format is tool-agnostic) | | A simple, open format specification for per-repo markdown files that guide AI coding agents across all tools — README.md for agents. |
| cmux (manaflow) cmux-manaflow · ui-dashboards | A12 UI passthrough | Tier B | 19,779 | 1 | claude-code | CLI UI | Native macOS terminal app with agent-aware notification rings, sidebar metadata per workspace, and scriptable browser — providing visibility into parallel coding agent sessions without prescribing workflow. |
| agentskills.io — Agent Skills Open Standard asyncope-agentskills-io · skill-catalogs | A13 Methodology | Tier B | 19,405 | 0.5 | claude-code | | Open standard specification for portable AI agent skill packages, enabling one skill to work across Claude Code, Gemini CLI, Codex, and any compliant agent. |
| Parlant parlant · linters | A11 Governance | Tier B | 18,084 | 4.1 | any-llm | CLI | Runtime conversation governance for customer-facing AI agents: per-turn contextual matching assembles only relevant behavioral rules into each LLM prompt, preventing attention failure at scale. |
| PydanticAI pydantic-ai · agent-sdks | A6 Multi-agent orchestrator | Tier B | 17,322 | 5.4 | any | CLI UI | Bring FastAPI's ergonomics (typed dependency injection, Pydantic validation, IDE auto-completion) to Python agent development. |
| Agent Skills for Context Engineering muratcankoylan-skills · skill-catalogs | A10 Memory engine | Tier B | 16,030 | 0 | claude-code | | 15-skill educational library covering the complete stack of context engineering and agent harness design — from attention mechanics to BDI cognitive modeling. |
| Eigent eigent · ui-dashboards | A12 UI passthrough | Tier B | 14,121 | 3 | own-runtime | UI MCP | Desktop cowork platform that deploys a multi-agent AI workforce (Developer, Browser, Document, Multi-Modal) to autonomously execute complex parallel tasks with human-in-the-loop checkpoints. |
| Browser Harness browser-harness · sandboxes | A18 Self-evolving | Tier B | 13,821 | 2.3 | Claude Code | CLI | Thin, self-healing CDP harness connecting an LLM to the user's real Chrome browser with coordinate-first clicking and agent-written helper code that improves itself each run. |
| OpenMythos openmythos · agent-sdks | A4 Markdown scaffold | Tier B | 13,396 | 0 | unknown | | Open-source PyTorch implementation of a theoretical Recurrent-Depth Transformer (RDT) architecture inspired by speculation about the Claude Mythos architecture. |
| Microsoft Agent Framework ms-agent-framework · agent-sdks | A8 Cross-runtime harness | Tier B | 10,763 | 5 | Azure AI Foundry | CLI UI | Production-grade Python/.NET SDK for building, orchestrating, and hosting multi-agent AI workflows on Azure Foundry. |
| InsForge insforge · agent-sdks | A10 Memory engine | Tier B | 10,636 | 1.7 | MCP server | UI MCP | Gives coding agents direct access to backend infrastructure (database, auth, storage, functions, model gateway) via MCP tools. |
| Claude Code System Prompts claude-code-system-prompts · standalone-clis | A13 Methodology | Tier B | 10,516 | 0 | claude-code | | Reverse-engineered reference of all 220+ internal Claude Code system prompts, tool descriptions, and agent instructions, tracked across 187 versions. |
| Hive (aden-hive) hive-aden · agent-sdks | A6 Multi-agent orchestrator | Tier B | 10,430 | 6.5 | claude-code | CLI UI MCP | Auto-generate and execute self-healing multi-agent DAGs for complex business workflows from a natural language goal. |
| BehiSecc/awesome-claude-skills behisecc-claude-skills · skill-catalogs | A13 Methodology | Tier B | 9,246 | 0 | unknown | | Curated awesome-list index of 50+ Claude Skills across 14 categories — a discovery surface, not a framework. |
| VoltAgent voltagent · agent-sdks | A6 Multi-agent orchestrator | Tier B | 9,175 | 6.4 | node | CLI UI MCP | TypeScript-native AI agent framework with typed contracts, composable guardrails, and a cloud ops console. |
| CAI (Cybersecurity AI) cai-cybersecurity · agent-sdks | A8 Cross-runtime harness | Tier B | 8,753 | 5.1 | openai-compatible | CLI UI | Lightweight framework for AI-powered offensive and defensive security automation, battle-tested in CTF competitions and real-world vulnerability discovery. |
| mcp-agent mcp-agent-lastmile · agent-sdks | A6 Multi-agent orchestrator | Tier B | 8,341 | 4.5 | any | CLI | Build durable, composable agents using MCP as the only protocol, implementing every pattern from Anthropic's Building Effective Agents paper. |
| Yao yao · agent-sdks | A6 Multi-agent orchestrator | Tier B | 7,541 | 2.5 | any | CLI UI MCP | Run AI agents and full-stack web apps from a single Go binary with TypeScript hooks, container sandboxing, and GraphRAG — zero runtime dependencies. |
| Plano plano · linters | A8 Cross-runtime harness | Tier B | 6,543 | 3.7 | any-http-agent | CLI | AI-native proxy built on Envoy that externalizes agent orchestration, LLM routing, observability, and safety as out-of-process middleware — agents are just HTTP servers. |
| OpenShell openshell-nvidia · sandboxes | A9 Sandbox substrate | Tier B | 6,264 | 4.7 | Claude Code | CLI UI | Safe, policy-governed sandbox runtime for autonomous AI agents with four-layer defense-in-depth (filesystem/network/process/inference) and hot-reloadable YAML policies. |
| holaOS hola-os · ui-dashboards | A12 UI passthrough | Tier B | 5,418 | 4.5 | own-runtime | UI | macOS desktop agent environment where recurring work-streams persist as living workspaces with accumulated user-correction rules, enabling compounding improvement across runs. |
| Cloudflare Agents cloudflare-agents · agent-sdks | A6 Multi-agent orchestrator | Tier B | 4,984 | 6.4 | Cloudflare Workers | CLI UI | Build persistent, stateful AI agents backed by Cloudflare Durable Objects — scale to millions of instances, each hibernating at zero cost when idle. |
| agent-infra sandbox agent-infra-sandbox · sandboxes | A9 Sandbox substrate | Tier B | 4,809 | 1 | any | UI MCP | All-in-one Docker container sandbox combining Browser (VNC), Terminal, File, VSCode Server, Jupyter, and MCP in a unified environment with shared filesystem. |
| Claude Code Ultimate Guide claude-code-ultimate-guide · methodology-docs | A13 Methodology | Tier B | 4,500 | 2.1 | claude-code | MCP | Comprehensive educational reference for Claude Code with 28-CVE security database, 271-question quiz, 181 templates, and companion MCP server for content navigation. |
| OpenAgentsControl (OAC) openagentscontrol · linters | A6 Multi-agent orchestrator | Tier B | 4,153 | 0.9 | opencode-cli | CLI | AI agents generate code matching your team's patterns by loading project context before every execution. |
| directories directories-leerob · skill-catalogs | A4 Markdown scaffold | Tier B | 3,936 | 0 | cursor | UI | Community plugin directory for Cursor IDE (cursor.directory) — redirected from leerob/directories; not a multi-model routing framework |
| ContextForge contextforge · mcp-tools | A11 Governance | Tier B | 3,770 | 3.2 | any-mcp-client | CLI UI MCP | Federates MCP servers, REST APIs, and A2A agents behind one authenticated endpoint with governance plugins and OpenTelemetry observability. |
| Archestra archestra · mcp-tools | A8 Cross-runtime harness | Tier B | 3,745 | 6.5 | any-mcp-client | UI | Enterprise AI platform providing centralized MCP registry, Kubernetes-native orchestration, dual-LLM security, and cost management for organizations adopting AI tools at scale. |
| LangChain MCP Adapters langchain-mcp-adapters · agent-sdks | A6 Multi-agent orchestrator | Tier B | 3,540 | 0 | LangGraph | | Bridge MCP tool protocol to LangChain/LangGraph BaseTool interface with zero boilerplate. |
| Open Claude Cowork open-claude-cowork · ui-dashboards | A12 UI passthrough | Tier B | 3,290 | 0 | claude-agent-sdk | UI | Minimal Electron desktop GUI for autonomous task execution via the Anthropic Claude Agent SDK — no terminal required. |
| Microsoft MCP Servers ms-mcp-servers · skill-catalogs | A3 MCP-anchored | Tier B | 3,209 | 2.9 | vscode-copilot | MCP | Official Microsoft MCP server catalog and source for Azure MCP and Fabric MCP servers, exposing 40+ Azure services through a single MCP server for any AI client. |
| Docker Agent docker-agent · agent-sdks | A6 Multi-agent orchestrator | Tier B | 2,962 | 2.5 | any | CLI UI | Define multi-agent AI systems in YAML, package via OCI, and run with Docker — no code required. |
| AgentGateway agentgateway · mcp-tools | A11 Governance | Tier B | 2,871 | 3.7 | any-protocol-client | CLI UI | Rust-based agentic proxy providing drop-in security, observability, and governance for MCP, A2A, and LLM traffic across any AI framework. |
| Amazon Bedrock AgentCore Samples agentcore-samples · skill-catalogs | A6 Multi-agent orchestrator | Tier B | 2,868 | 6.9 | strands-agents | CLI MCP | Managed cloud infrastructure for deploying AI agents at production scale — eliminating undifferentiated heavy lifting of compute, memory, identity, and security. |
| Google Agents CLI google-agents-cli · skill-catalogs | A6 Multi-agent orchestrator | Tier B | 2,589 | 5.4 | gemini-cli | CLI | Google's official CLI and skill pack for building production ADK agents — enforcing scaffold-first, eval-driven development across Claude Code, Gemini CLI, Codex, and any compliant coding agent. |
| Agent Deck agent-deck · ui-dashboards | A6 Multi-agent orchestrator | Tier B | 2,535 | 4.5 | claude-code | CLI UI | Terminal mission control for managing fleets of parallel AI coding agent sessions with fork, MCP, conductor, and worktree support. |
| CCS (Claude Code Switch) ccs-switch · standalone-clis | A16 Cross-vendor router | Tier B | 2,410 | 2.6 | claude-code | CLI UI | Routes Claude Code tasks to the optimal AI agent (Codex, Factory Droid, Ollama, Claude) based on configurable cost/capability profiles, with OAuth account pool management to avoid rate limits and a web dashboard for usage analytics. |
| NeMo Agent Toolkit nemo-agent-toolkit · agent-sdks | A6 Multi-agent orchestrator | Tier B | 2,333 | 3.6 | any | CLI UI MCP | Add enterprise-grade instrumentation, profiling, evaluation, and RL fine-tuning to any AI agent framework without migrating your code. |
| cmux (coder) / Mux cmux-coder · ui-dashboards | A11 Governance | Tier B | 1,782 | 6.7 | mux-own-agent-loop | CLI UI MCP | Desktop app with custom agent loop for parallel, isolated, multi-model agentic development with Plan/Exec modes, worktree isolation, and sub-agent code review. |
| Microsoft Learn MCP Server ms-learn-mcp · skill-catalogs | A3 MCP-anchored | Tier B | 1,661 | 1.3 | vscode-copilot | CLI MCP | Free, no-auth MCP server providing real-time access to official Microsoft documentation, eliminating AI hallucinations about Azure APIs, .NET methods, and Microsoft SDKs. |
| stakpak/agent stakpak-agent · sandboxes | A9 Sandbox substrate | Tier B | 1,563 | 5 | stakpak-cli | CLI UI MCP | 24/7 DevOps AI agent with dynamic secret substitution, Warden pre-execution guardrails, and Docker sandbox isolation for safe autonomous infrastructure operations. |
| Scion (GCP) scion-gcp · sandboxes | A16 Cross-vendor router | Tier B | 1,548 | 6.5 | gemini-cli | CLI UI | Multi-agent orchestration testbed: each agent in own container + git worktree, harness-agnostic (Claude/Gemini/Codex/OpenCode), OTEL telemetry, emergent coordination via CLI tool teaching. |
| Code-Mode Library code-mode-library · mcp-tools | A19 Token-efficient encoding | Tier B | 1,461 | 1.5 | claude-code | MCP | Replaces traditional tool-calling with TypeScript code execution in a sandbox, collapsing N sequential tool calls into 1 code execution with all registered tools available as typed namespace functions. |
| E2B Desktop Sandbox e2b-desktop · sandboxes | A9 Sandbox substrate | Tier B | 1,388 | 1 | any | | Cloud-managed graphical desktop sandbox (MicroVM + VNC) for LLM Computer Use agents — create, control, and stream a real GUI environment via simple SDK. |
| ClawManager clawmanager · sandboxes | A11 Governance | Tier B | 1,351 | 5 | openclaw | UI | Kubernetes-native control plane for multi-tenant AI agent instance management with AI Gateway governance, desired-state orchestration, and skill/channel resource management. |
| AiderDesk aiderdesk · ui-dashboards | A12 UI passthrough | Tier B | 1,233 | 4.7 | aider-cli | UI MCP | Transparent, steerable AI orchestration platform — GUI for Aider with multi-provider support, worktree isolation, approval gates, and a TypeScript extension ecosystem. |
| AgentBay SDK agentbay-sdk · sandboxes | A8 Cross-runtime harness | Tier B | 1,124 | 3.2 | any | | Multi-language SDK for Alibaba Cloud's on-demand sandbox sessions across Browser, Desktop, Mobile, and Code execution surfaces. |
| Caliber caliber · standalone-clis | A8 Cross-runtime harness | Tier B | 1,087 | 4.4 | claude-code | CLI | Generates and continuously maintains AI context files (CLAUDE.md, Cursor rules, AGENTS.md, Copilot instructions) with deterministic quality scoring and automatic refresh on every commit. |
| codesight codesight · standalone-clis | A3 MCP-anchored | Tier B | 1,074 | 2.2 | claude-code | CLI UI MCP | Pre-compiles codebase structure via AST analysis into minimal-token context maps so AI assistants spend zero tokens exploring project structure at session start. |
| Mysti mysti · ui-dashboards | A8 Cross-runtime harness | Tier B | 1,056 | 5.2 | claude-code | UI | VS Code extension that orchestrates 12 AI coding agent CLIs through a unified chat UI with 5-strategy Brainstorm Mode, 16 personas, and heuristic convergence detection for multi-agent collaboration. |
| Sidecar sidecar-marcus · ui-dashboards | A8 Cross-runtime harness | Tier B | 1,018 | 2 | claude-code | CLI UI | Terminal TUI companion panel that provides real-time git diff viewing, unified conversation history from 10+ AI coding agents, task monitoring, and workspace management alongside any coding agent session. |
| Tensorlake tensorlake · sandboxes | A9 Sandbox substrate | Tier B | 926 | 6.2 | any | CLI | Fastest Firecracker MicroVM sandbox platform with sub-second cold starts, runtime snapshots/clones, and integrated serverless fan-out orchestration for Python agent pipelines. |
| Awesome Claude Code Plugins ccplugins-awesome · skill-catalogs | A2 Mirror cmd+skill | Tier B | 806 | 1.6 | claude-code | | Community-curated marketplace of 116 installable Claude Code plugins covering the full developer workflow, from code review to business automation. |
| AgentScope Runtime agentscope-runtime · sandboxes | A9 Sandbox substrate | Tier B | 800 | 4.5 | AgentScope | UI | Production runtime for exposing any Python agent as a scalable HTTP/SSE streaming service with containerized tool execution sandbox. |
| TraceRoot AI traceroot-ai · linters | A10 Memory engine | Tier B | 591 | 5.1 | standalone-repo | UI MCP | Automatically detect AI agent failures in production via LLM-as-judge monitoring and AI-powered root cause analysis linked to source code and git history. |
| Claude Skills Marketplace (mhattingpete) claude-skills-marketplace-mhatting · skill-catalogs | A19 Token-efficient encoding | Tier B | 586 | 0.5 | claude-code | MCP | Curated multi-plugin marketplace for SW engineering workflows with optional token-efficient code execution runtime for bulk file operations. |
| MyCoder.ai mycoder-ai · coding-agents | A8 Cross-runtime harness | Tier B | 567 | 1.5 | standalone-cli | CLI | Autonomous parallel-executing coding agent that reads project context naturally without special setup files. |
| SwarmVault swarmvault · memory-tools | A18 Self-evolving | Tier B | 492 | 4.1 | Claude Code | CLI UI MCP | Production-grade CLI for Karpathy's LLM Wiki pattern: ingests any content into a local-first durable markdown wiki + knowledge graph with MCP server, approval queues, and bounded agent handoff context packs. |
| Vibe Check MCP vibe-check-mcp · mcp-tools | A3 MCP-anchored | Tier B | 486 | 3.7 | claude-desktop | CLI MCP | Provides a second LLM as a meta-mentor that interrupts agent pattern inertia with Chain-Pattern Interrupts before irreversible actions or when reasoning may be locked-in. |
| Disciplined-AI-Software-Development (Varietyz) brian-disciplined-ai · methodology-docs | A13 Methodology | Tier B | 402 | 2 | any | | Four-stage methodology + PAG instruction syntax + 5 personas to enforce behavioral consistency across web browser and CLI AI sessions — prevents vibe coding via measurable architectural constraints. |
| Claude Skills MCP Server claude-skills-mcp · skill-catalogs | A3 MCP-anchored | Tier B | 387 | 0 | cursor | CLI MCP | Brings Anthropic's Agent Skills semantic discovery to any MCP-compatible AI client via a two-package proxy+backend architecture. |
| IBM MCP ibm-mcp · skill-catalogs | A3 MCP-anchored | Tier B | 379 | 1 | vscode-copilot | MCP | IBM's official catalog of 30+ MCP servers connecting AI agents to the full IBM product portfolio — from watsonx.data and QRadar to z/OS mainframe and IBM i. |
| AgentOps (boshu2) agentops-boshu · linters | A8 Cross-runtime harness | Tier B | 369 | 9.3 | claude-code | CLI | SDLC control plane for coding agents: compounding corpus of decisions, learnings, and planning rules that makes each session smarter than the last. |
| Subtask subtask-zippoxer · skill-catalogs | A6 Multi-agent orchestrator | Tier B | 330 | 7 | claude-code | CLI UI | Gives a lead AI agent a Go CLI + git worktree infrastructure to dispatch parallel worker agents, track progress, and orchestrate merges without user involvement. |
| aws-skills aws-skills · skill-catalogs | A1 Skills-only | Tier B | 294 | 1.5 | claude-code | MCP | Gives Claude Code mandatory-MCP-first AWS knowledge for CDK, serverless, cost optimization, and Bedrock AgentCore through 5 domain skills with bundled MCP server configurations. |
| claude-teams (cs50victor) claude-teams-mcp · mcp-tools | A3 MCP-anchored | Tier B | 271 | 4.5 | claude-code | UI MCP | Reimplements Claude Code's internal agent teams protocol as a standalone MCP server, enabling cross-agent multi-agent coordination between Claude Code and OpenCode. |
| curtis-arch/Droids droids-curtis · skill-catalogs | A16 Cross-vendor router | Tier B | 253 | 1 | factory-ai-droid-cli | | 29 production-tested, hyper-specific domain-specialist droids for a Cloudflare Workers/Neon/tRPC stack, with forced context injection replacing codebase exploration. |
| CCBot ccbot · ui-dashboards | A4 Markdown scaffold | Tier B | 250 | 2.3 | claude-code | CLI | Telegram bot that bridges to tmux-based Claude Code sessions, allowing remote monitoring and control via Telegram topics without interrupting the terminal session. |
| claude-context-local claude-context-local · memory-tools | A10 Memory engine | Tier B | 231 | 1 | claude-code | MCP | Provides semantic code search via local EmbeddingGemma model and FAISS, giving Claude Code precise code chunk retrieval without cloud API costs or data transmission. |
| glebis/claude-skills glebis-skills · skill-catalogs | A6 Multi-agent orchestrator | Tier B | 225 | 6.5 | claude-code | CLI | 48-plugin personal productivity OS with flagship multi-agent TDD skill implementing information isolation between test-writing and implementation phases. |
| grov grov · memory-tools | A10 Memory engine | Tier B | 192 | 4.2 | claude-code | CLI UI MCP | Share what one developer's AI agent learns automatically with all team members' agents, eliminating redundant exploration and duplicate token spend. |
| Gemini CLI Skill for Claude Code gemini-cli-skill-foray · mcp-tools | A6 Multi-agent orchestrator | Tier B | 184 | 1.5 | claude-code | | Teaches Claude Code to delegate tasks to Gemini 3 Pro via CLI for second opinions, Google Search grounding, and codebase architecture analysis. |
| Poirot poirot · ui-dashboards | A12 UI passthrough | Tier B | 184 | 0 | claude-code | UI | Native macOS session investigator — browse, search, and analyze Claude Code history without modifying any workflow. |
| agentbox agentbox-mattolson · sandboxes | A16 Cross-vendor router | Tier B | 174 | 3 | claude-code | CLI | Two-layer network-isolated local sandbox (mitmproxy allowlist + iptables firewall) for running AI coding agents with minimal filesystem access and proxy-injected credentials. |
| ContextKit contextkit · standalone-clis | A6 Multi-agent orchestrator | Tier B | 169 | 3 | claude-code | | Spec-driven 4-phase planning workflow for iOS/macOS Swift development in Claude Code, with quality-specialized subagents for accessibility, localization, and modern code patterns. |
| Pare pare-mcp · mcp-tools | A19 Token-efficient encoding | Tier B | 122 | 0.6 | claude-code | CLI MCP | 28 MCP servers providing schema-validated JSON output for 240 developer tools, eliminating ANSI escape codes and fragile CLI output parsing with 36-95% token reduction. |
| KubeStellar Console kubestellar-kc-agent · sandboxes | A10 Memory engine | Tier B | 109 | 5.6 | Claude Code | CLI UI MCP | AI-powered multi-cluster Kubernetes management dashboard with guided install missions for 250+ CNCF projects, accessible as hosted demo or self-hosted with local kc-agent bridge. |
| IM.codes im-codes · ui-dashboards | A10 Memory engine | Tier B | 100 | 5 | claude-code | CLI UI | Remote control and shared memory layer for terminal-based AI coding agents, accessible from mobile, web, and Apple Watch. |
| osModa osmoda · ides | A20 OS-level substrate | Tier B | 99 | 7.5 | openclaw | CLI UI MCP | AI-native NixOS operating system where 10 Rust daemons expose 91 structured MCP tools giving the agent root-level typed access to the entire system, with atomic rollback on every change and a SHA-256 hash-chained tamper-proof audit ledger. |
| terminal-bench-env (TermiGen) terminal-bench-env · sandboxes | A8 Cross-runtime harness | Tier B | 82 | 2.4 | none (model-agnostic) | | 3,500+ verified Docker terminal tasks + minimal ReAct BashAgent for evaluating and training terminal-capable AI agents. |
| AGENT.md agent-md · methodology-docs | A13 Methodology | Tier B | 80 | 0.5 | amp (native support) | | RFC-style competing format standard to AGENTS.md: one universal AGENT.md file replaces tool-specific config files (.cursorrules, CLAUDE.md, .windsurfrules) via symbolic linking. |
| onUI onui · ui-dashboards | A3 MCP-anchored | Tier B | 62 | 0.5 | claude-code | CLI UI MCP | Browser extension + local MCP server that lets developers annotate any live web UI with structured intent/severity markers, exposing those annotations to AI coding agents as queryable context without modifying app source code. |
| skill-optimizer skill-optimizer · skill-catalogs | A8 Cross-runtime harness | Tier B | 57 | 5.1 | claude-code | CLI | Runs AI agent skills in hermetically isolated Docker containers against deterministic graders to produce reproducible pass/fail results across a model × case matrix. |
| xiaolai/nlpm-for-claude nlpm-xiaolai · linters | A18 Self-evolving | Tier B | 55 | 4.1 | claude-code | CLI MCP | Treats NL artifacts (skills, agents, commands, CLAUDE.md) as lintable programs with a 50-rule rubric, penalty-based scoring, NL-TDD, and a self-evolving exemplar pipeline. |
| Vibe Island vibe-island · ui-dashboards | A12 UI passthrough | Tier B | 54 | 0 | claude-code | UI | macOS notch panel showing real-time status for 16 AI coding agents with GUI permission approval and one-click terminal jump. |
| mcp2skill mcp2skill · mcp-tools | A8 Cross-runtime harness | Tier B | 51 | 2 | claude-code | CLI | Converts any MCP server into a production-quality agent skill package following the agentskills.io specification, with real introspection, OAuth support, and update detection. |
| AgentTrace agenttrace-luoyu · linters | A8 Cross-runtime harness | Tier B | 49 | 1.2 | claude-code | CLI UI | Local-first TUI and CI-gate tool for retrospective analysis of AI coding-agent session cost, tokens, health, and latency across 15+ agent runtimes. |
| Heavy3 Code Audit heavy3-code-audit · linters | A16 Cross-vendor router | Tier B | 44 | 1.9 | claude-code | | Routes code and plan review through a 3-model council (GPT/Gemini/Grok) to catch what any single model misses via multi-model consensus. |
| unslop unslop · standalone-clis | A8 Cross-runtime harness | Tier B | 44 | 0.9 | claude-code | CLI | Strips AI-isms (sycophancy, stock vocabulary, hedging stacks, em-dash overuse) from LLM responses while preserving technical accuracy via research-backed subtraction rules. |
| LynxPrompt lynxprompt-geiserx · standalone-clis | A8 Cross-runtime harness | Tier B | 41 | 1.7 | claude-code | CLI UI MCP | Self-hostable platform for creating, versioning, and distributing AI coding assistant configurations (CLAUDE.md, AGENTS.md, .cursorrules) across teams with enterprise auth, federated marketplace, and 30+ format export. |
| Agent FM agent-fm · ui-dashboards | A11 Governance | Tier B | 40 | 1 | claude-code | UI | Ambient audio radio station for AI coding agent sessions — narrates progress, blockers, and decisions without terminal attention. |
| Aurite Agent Verifier aurite-agent-verifier · skill-catalogs | A8 Cross-runtime harness | Tier B | 38 | 0.4 | claude-code | | Applies deterministic pattern-matched checks on AI-generated agent code to catch hallucinated tools, unbounded loops, and missing retry limits before code ships. |
| awesome-gemini-cli-subagents ankitmundada-gemini-subagents · skill-catalogs | A11 Governance | Tier B | 33 | 1.6 | gemini-cli | | Provide 128 specialized persona-MD agents for Gemini CLI with per-agent model routing and temperature optimization. |
| claude-code-mcp (KunihiroS) claude-code-mcp-kunihiros · mcp-tools | A10 Memory engine | Tier B | 33 | 2.2 | claude-desktop | CLI MCP | Expose Claude Code CLI capabilities as MCP tools accessible from any MCP host application. |
| Twelve-Factor Agentic SDLC agentic-sdlc-12-factors · methodology-docs | A13 Methodology | Tier B | 29 | 3.7 | unknown | | Twelve prescriptive factors for integrating AI coding agents into team SDLC — addressing coordination failures, knowledge silos, and governance gaps that individual AI productivity gains don't solve. |
| faf-cli faf-cli · standalone-clis | A8 Cross-runtime harness | Tier B | 29 | 1.7 | claude-code | CLI UI | Provides a git-native, IANA-registered YAML format for AI project context (project.faf) that bi-syncs with CLAUDE.md/AGENTS.md and scores 0-100% AI-readiness via a WASM kernel. |
| Agentify agentify · standalone-clis | A8 Cross-runtime harness | Tier B | 28 | 1.9 | claude-code | CLI | Compiles OpenAPI specifications into all agent interface formats (MCP server, CLAUDE.md, AGENTS.md, skills, .cursorrules, A2A card, CLI) so API owners don't manually maintain 9 synchronized files as their API evolves. |
| OpenQuack openquack · ui-dashboards | A12 UI passthrough | Tier B | 28 | 0 | claude-code | UI | On-device voice dictation for macOS that converts speech to text at the cursor in any app, with roadmap integration for live AI coding agent sessions. |
| claude-code-playbook (RiyaParikh0112) claude-code-playbook-riya · methodology-docs | A13 Methodology | Tier B | 26 | 0.5 | claude-code | | Written guide teaching the paradigm shift from hands-on coder to AI development director using Claude Code. |
| Claude Code Plan Export claude-code-plan-export · memory-tools | A10 Memory engine | Tier B | 21 | 0.5 | claude-code | | Auto-export Claude Code Plan Mode plans to the project root on session end so they can be executed in a fresh context with a cheaper model. |
| 094459/sdd-workshop 094459-sdd-workshop · methodology-docs | A13 Methodology | Tier B | 18 | 1.5 | kiro | | Teaches developers the Kiro IDE's spec-driven development workflow through hands-on labs for both greenfield and brownfield scenarios. |
| prd-creator-skill huchi996-prd-creator · skill-catalogs | A4 Markdown scaffold | Tier B | 17 | 1.8 | claude-code | | Converts a vague Chinese-language product idea into a structured three-tier PRD hierarchy (Project→Module→Function) through a 6-phase AI-driven SOP with multi-role review. |
| Anima anima · ui-dashboards | A12 UI passthrough | Tier B | 15 | 0 | claude-code | UI MCP | Gamified macOS companion replacing terminal for Claude Code — cross-session watcher, ASCII familiars, nim tokens, voice input, oracle commentary. |
| setup-structure-index setup-structure-index · skill-catalogs | A10 Memory engine | Tier B | 15 | 1.5 | claude-code | | Creates a two-tier codebase index (zero-cost file map in CLAUDE.md + on-demand directory YAML) maintained by a TaskCompleted hook that fires only on structural changes. |
| Arbor arbor-zed-gpui · linters | A4 Markdown scaffold | Tier B | 14 | 0 | other | | Static API gateway framework (NOT an AI agent framework — slug mismatch in batch assignment). |
| oh-my-claude (ssenart) ssenart-oh-my-claude · skill-catalogs | A4 Markdown scaffold | Tier B | 14 | 0 | claude-code | UI | Provides a real-time powerline status bar in Claude Code showing context usage, token count, Pro rate limit percentages, and reset timers. |
| Claude Tools MCP Server claude-tools-mcp · mcp-tools | A3 MCP-anchored | Tier B | 13 | 0.8 | any-mcp-client | CLI MCP | Exposes Claude Code's file and shell tools (bash, read, write, edit, glob, grep) as an MCP server for remote access by any MCP client. |
| sgaunet/claude-plugins sgaunet-plugins · skill-catalogs | A6 Multi-agent orchestrator | Tier B | 13 | 1.9 | claude-code | MCP | Four-plugin domain-specialist marketplace (DevOps, software engineering, Go, Bash) with named expert agents and per-agent tool restrictions. |
| huangdijia/oh-my-claude-code-plugins huangdijia-cc-plugins · skill-catalogs | A6 Multi-agent orchestrator | Tier B | 10 | 1.5 | claude-code | MCP | Multi-plugin Claude Code marketplace implementing spec-driven development (EARS requirements + Mermaid design + task checklist) with surgical one-at-a-time task execution and git automation hooks. |
| House MCP Manager house-mcp-manager · mcp-tools | A8 Cross-runtime harness | Tier B | 8 | 2.2 | claude-code | CLI UI | Manages MCP server enable/disable state across Claude Code, Cursor, and Cline to prevent context window exhaustion from startup tool-loading. |
| Mantra mantra · ui-dashboards | A12 UI passthrough | Tier B | 7 | 0 | claude-code | UI | Session time machine — maps each conversation turn to exact filesystem state for time-travel debugging and session replay. |
| anilcancakir/claude-code-plugins anilcancakir-cc-plugins · skill-catalogs | A6 Multi-agent orchestrator | Tier B | 6 | 1.5 | claude-code | | Seven productivity plugins for Claude Code with strategic-compact as a hook-driven context lifecycle manager that suggests compaction at phase transitions rather than arbitrary auto-compact points. |
| codebase-pilot-cli codebase-pilot-cli · standalone-clis | A8 Cross-runtime harness | Tier B | 6 | 5.4 | claude-code | CLI UI MCP | Combines codebase context packing, security scanning, workflow skills, and a persistent web dashboard into one zero-cloud tool that reduces per-prompt token cost by 60-90% while tracking savings over time. |
| PromptKit (ozzeron/prompt-pack) ozzeron-prompt-pack-promptkit · skill-catalogs | A8 Cross-runtime harness | Tier B | 6 | 0.5 | cursor | | Prevents AI agents from creating duplicate artifacts by enforcing a mandatory reuse-before-create decision chain as a precondition for any code-writing action. |
| claude-code-playbook (toy-crane) claude-code-playbook-toy-crane · methodology-docs | A6 Multi-agent orchestrator | Tier B | 3 | 4.5 | claude-code | UI | Korean-language Claude Code lecture site with QA reviewer agents and student plugin. |
| agentic-python-coder szeider-python-coder · coding-agents | A8 Cross-runtime harness | Tier B | 3 | 5.2 | standalone | CLI MCP | A focused Python execution agent (LangGraph ReAct + persistent IPython kernel) that mandates constraint-independent verification before saving any solution, plus an MCP server for multi-kernel parallel Python execution. |
| claude-code-playbook (avinashmyerolkar) claude-code-playbook-avinash · methodology-docs | A13 Methodology | Tier B | 2 | 0.5 | claude-code | | Personal study notes documenting Claude Code features with emphasis on spec-driven development workflow. |
| claude-overlay claude-overlay · standalone-clis | A4 Markdown scaffold | Tier B | 2 | 0 | claude-code | CLI | Manages project-level Claude Code configuration for non-Anthropic model endpoints, surgically injecting/removing only overlay-managed keys while preserving all user-custom settings, and replacing broken native web tools with MCP-based alternatives. |
| claude-snapshot claude-snapshot · standalone-clis | A4 Markdown scaffold | Tier B | 2 | 4 | claude-code | CLI | Exports and restores the complete Claude Code configuration (~/.claude/) as a portable, diff-inspectable archive for multi-machine sync, OS reinstall recovery, and team onboarding. |
| prpack prpack · standalone-clis | A8 Cross-runtime harness | Tier B | 2 | 0.5 | claude | CLI | Packages PR diffs with full post-change file contents so LLMs can review code with context about unchanged lines, not just what changed. |
| AI Context Linter ai-context-linter · linters | A8 Cross-runtime harness | Tier B | 1 | 0.4 | claude-code | | CI linter that validates AI coding context files (CLAUDE.md, .cursorrules, AGENTS.md) against 12 rules covering security, structure, and AI anti-patterns. |
| AI Context Templates ai-context-templates · methodology-docs | A8 Cross-runtime harness | Tier B | 1 | 0 | claude-code | UI | Provides annotated fill-in-the-blank starter files for AI coding context (CLAUDE.md, .cursorrules, PRP) so developers don't write useless or harmful context files. |
| oh-my-agent-skills lukasdias-oh-my-agent-skills · skill-catalogs | A8 Cross-runtime harness | Tier B | 1 | 0 | claude-code | CLI UI | Rust TUI for browsing, searching, filtering, and copying skill names from .agents/ skill directories used by Claude/Opencode coding assistants. |
| Anthropic Claude Plugins Official anthropics-plugins-official · skill-catalogs | A99 Unclassified | Tier B | 0 | 0 | unknown | | Repo not found — no public content to analyze. |
| Deebo deebo · mcp-tools | A3 MCP-anchored | Tier B | 0 | 5 | claude-code | MCP | Autonomous debugging MCP server that spawns parallel scenario agents to test competing bug hypotheses in isolated git branches, synthesizing a validated solution. |
| Marmot marmot · standalone-clis | A8 Cross-runtime harness | Tier B | 0 | 3.7 | claude-code | CLI UI | Provides a single shell-pipeable CLI for AI generation, web search, and data enrichment across 25+ providers so agents can delegate external calls without burdening the main context window. |
| mcp-to-skill-converter mcp-to-skill-converter · mcp-tools | A4 Markdown scaffold | Tier B | 0 | 1.5 | claude-code | | Converts any MCP server into a Claude Code skill using progressive disclosure to achieve ~90% context savings at idle. |
| acpx acpx · mcp-tools | A4 Markdown scaffold | Tier B | — | 0 | unknown | | |
| bringyour-migration-auditor asyncope-bringyour · linters | A4 Markdown scaffold | Tier B | — | 0 | other | | unknown — repository returns HTTP 404 |
| CCHub cchub · ui-dashboards | A4 Markdown scaffold | Tier B | — | 0 | unknown | | CC ecosystem management platform with visual GUI (insufficient public material to verify). |
| GitAgentProtocol (OpenGAP) gitagent-protocol · methodology-docs | A4 Markdown scaffold | Tier B | — | 0 | unknown | | |
| LIT.AI methodology lit-ai-methodology · methodology-docs | A13 Methodology | Tier B | — | 0 | unknown | | Blog-published software development methodology — no packaged framework available. |
| OmX (oh-my-codex-yeachan duplicate) oh-my-codex-yeachan-omx · skill-catalogs | A4 Markdown scaffold | Tier B | — | 0 | other | | Duplicate entry — see canonical oh-my-codex-yeachan |
| Onepilot onepilot · ui-dashboards | A12 UI passthrough | Tier B | — | 2.5 | openclaw | UI | Mobile iPhone app for deploying and controlling terminal AI coding agents (OpenClaw, Hermes, Claude Code) on any SSH-accessible host, with configurable agent personalities and Telegram/Discord/Slack messaging channels. |
| PATAPIM patapim · ui-dashboards | A12 UI passthrough | Tier B | — | 0 | claude-code | UI | Browser-automation-based Claude Code companion that uses Claude.ai subscriptions instead of API to eliminate per-token costs. |
| Ptah (Coding Orchestra) ptah-orchestra · ui-dashboards | A6 Multi-agent orchestrator | Tier B | — | 3.6 | unknown | CLI UI | |
| sightglass sightglass · linters | A4 Markdown scaffold | Tier B | — | 0 | unknown | | unknown — repository not accessible |