Skip to content
/
§ a4

Markdown scaffold (no primitives)

Ships structured markdown the agent reads passively. No commands, no skills, no hooks.

46 primary frameworks · 54 lower-confidence entries

BMAD-METHOD
bmad-method
★ 48k
Seed

Provides a full agile delivery lifecycle with named expert-persona AI collaborators that elicit the human's best thinking rather than substituting AI opinion…

Anthropic Knowledge Work Plugins
anthropics-knowledge-work-plugins
★ 16k
Tier A

Role-specialized plugin bundles with live MCP connectors that turn Claude into a domain expert for enterprise knowledge workers.

Agent OS
agent-os
★ 4.6k
Seed

Extracts implicit codebase conventions into token-efficient markdown standards files and injects them selectively into AI agent context windows so agents build…

Codex Integration for Claude Code (skill-codex)
codex-integration-skill
★ 1.3k
Tier A

Single Claude Code skill that handles Codex CLI invocation correctly (stdin blocking, thinking token suppression, session resume) with a peer-AI critical…

mini-coding-agent
mini-coding-agent
★ 882
Tier A

A single-file zero-dependency Python coding agent that demonstrates the six core components of coding agents for educational purposes, running entirely on…

beads-ui
beads-ui
★ 621
Tier A

Give human developers a browser-based kanban view of Beads issues while AI agents work in the CLI.

TheAuditor
theauditor
★ 550
Tier A

AI context and enforcement layer for code-generating agents (pre-launch; no public source).

Bernstein (sipyourdrink-ltd)
bernstein-sipyourdrink
★ 460
Tier A

Alias entry — same repo as bernstein slug. See spec-system/bernstein/ for full analysis.

Claude Conductor
claude-conductor
★ 367
Seed

Gives Claude Code a persistent, cross-linked, auto-analyzed documentation system so it retains codebase context across sessions.

Conductor (MadAppGang)
claude-conductor-madappgang
★ 280
Tier A

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.

Custom-Modes-Roo-Code (jtgsystems)
custom-modes-roo-code
★ 173
Tier A

225-agent configuration library for Roo Code covering every major software development domain with 2026 security-first standards.

Superpowers: Developing for Claude Code
superpowers-developing-for-claude-code
★ 129
Tier A

Gives Claude Code comprehensive, self-updating knowledge of the Claude Code platform for building plugins, skills, hooks, commands, and MCP servers.

Utah (Inngest)
utah-inngest
★ 116
Tier A

Durable Inngest-powered personal agent where every LLM call and tool execution is a checkpointed step with automatic retries, singleton concurrency, and…

fspec (SENGAC)
fspec
★ 69
Tier A

Provides complete ACDD infrastructure for AI agents: Gherkin spec management, Example Mapping facilitation, TDD state machine enforcement, Git checkpoints for…

VibeSec
vibesec
★ 46
Tier A

Injects OWASP-aligned security rules into Cursor and Windsurf AI assistants to prevent common security vulnerabilities in AI-generated code.

Uniswap AI Toolkit
ai-toolkit-uniswap
★ 37
Tier A

Organizational Claude Code plugin marketplace that packages Uniswap Labs' engineering culture (feature planning, PR workflows, codebase navigation,…

SuperBeads Universal Framework (Wiggum Flavour)
superbeads-wiggum
★ 31
Tier A

Bring Beads task tracking, atomic git commits, session memory, and domain-specific verification to Claude Code via a universal CLI-installed scaffold.

FlowCoder
flowcoder
★ 26
Tier A

Build complex multi-step autonomous agent workflows visually via a flowchart builder — no orchestration code required.

Spec-Driver (Greenfield Spec-Driven Development)
spec-driver
★ 25
Seed

Prevents spec rot in AI-assisted development by making implementation changes flow back into evergreen, authoritative specs via audits, drift ledgers, and…

Conductor (fcoury)
conductor-fcoury
★ 25
Tier A

Cross-tool CDD implementation running identical methodology on Claude Code, Gemini CLI, and Codex CLI with an auto-activating ambient skill.

roomodes (sellisd)
roomodes-sellisd
★ 23
Tier A

Simulate a full agile scrum team (PO, SM, Dev, QA) in Roo Code using per-role Jira-integrated personas.

Eval Marketplace
eval-marketplace
★ 22
Tier A

Automated supply-chain security evaluation of third-party agent skills and MCP servers before installation, using GitHub and web-scraping MCP tools.

CC Marketplace (setouchi-h)
cc-marketplace-setouchi
★ 16
Tier A

Automate the repetitive git, PR, and build tasks that surround every Claude Code session without adding workflow methodology.

ching-kuo/claude-codex
ching-kuo-claude-codex
★ 16
Tier A

Bidirectional Claude↔Codex critique loop via MCP where Claude plans/implements and Codex audits/reviews with APPROVED/WARNING/BLOCKED verdicts.

Vibook (pinqy520)
roo-code-vibook
★ 16
Tier A

Generate an AI-optimized documentation knowledge base for any existing codebase to enable efficient AI-assisted development and maintenance.

CCCT (IgorWarzocha)
ccct
★ 15
Tier A

Professional development workflow chain (setup→explore→PRD→tasks→dev→review) with complexity-driven task breakdown and dual-tracking for Claude Code.

Cursor Agentic Development Pipeline
cursor-agentic-pipeline
★ 12
Tier A

Automate building an Obsidian knowledge base of Laravel packages, skills, and MCP servers using GitHub Copilot cloud agents.

Agent Harness Skill (Harness Engineering)
ldzhouquan-agent-harness-skill
★ 9
Tier A

Systematic Harness Engineering methodology with Iron Laws, Reflexion Loop, and 5-second session-amnesia recovery protocol.

spec-driven-kiro-project (lotharschulz)
spec-driven-kiro-project
★ 3
Tier A

Example project demonstrating Kiro IDE's spec-driven development approach applied to a real TypeScript web application.

spec-kit-command-cursor (foxgod183)
spec-kit-command-cursor-foxgod
★ 3
Tier A

Simplified Cursor IDE spec-driven toolkit with 8 commands and compact template variants for non-technical users.

doodledood/codex-workflow
doodledood-codex-workflow
★ 2
Tier A

Full-lifecycle Codex CLI skill pack built around documented LLM limitations, with 8-type specialized review orchestration and dynamic requirements discovery.

vibelint
vibelint
★ 2
Tier A

Post-session linter that audits Claude Code session transcripts for quality, cost, and config issues.

Simple Agent Harness
bquast-simple-agent-harness
★ 0
Tier A

Minimal autonomous LLM agent: cron + Ollama + soul/memory/heartbeat markdown files, no dependencies.

cc-simple-memory (STRML)
cc-simple-memory
★ 0
Tier A

Minimal shell + Node.js plugin with 3 lifecycle hooks (SessionStart/Stop/PreCompact) that extracts session learnings into per-project and global Markdown files…

claudeAutoPilot
claudeautopilot
★ 0
Tier A

PRD-driven autonomous development framework that manages a 4-column Kanban workflow, runs quality gates, and surfaces only classified decisions…

roo-code-community (James-Cherished-Inc)
james-cherished-roo-community
★ 0
Tier A

Community hub for Roo Code users: one-command custom mode installation, Perplexity MCP setup, and a web UI for managing .roomodes configurations.

Nanocoder
nanocoder
★ 0
Tier A

unknown — repository not publicly accessible

spec-kit-aider (wolffy-au)
spec-kit-aider
★ 0
Tier A

Aider-adapted fork of GitHub spec-kit that adds automated numbered git branch creation and cross-platform scripts for spec-driven development.

AI Agents in Enterprise Product Development Spec
enterprise-ai-agents-spec
Tier A
FlyDex
flydex
Tier A

unknown — no public GitHub repository accessible

Headless Coder SDK
headless-coder-sdk
Tier A

unknown — repository not found (HTTP 404)

Jeffallan/claude-skills
jeffallan-claude-skills
Tier A
PAF Framework
paf-framework
Tier A

Repository not found — crack00r/paf-framework returns 404.

roocode-memorybank-optimized
roocode-memorybank-optimized
Tier A

unknown — repository not accessible

squad-framework
squad-framework
Tier A
VibeBox
vibebox
Tier A

unknown — no public GitHub repository accessible

Show 54 lower-confidence entriestier-b · tier-c · unknown · delta reports

These entries map to § a4 by tag but carry weaker evidence — fewer documented primitives, delta reports of absent skills, or marketing-only sites without a public repo. They're listed for completeness; treat them with appropriate caution.

OpenMythos ★ 13k
openmythos

Open-source PyTorch implementation of a theoretical Recurrent-Depth Transformer (RDT) architecture inspired by speculation about the Claude Mythos architecture.

directories ★ 3.9k
directories-leerob

Community plugin directory for Cursor IDE (cursor.directory) — redirected from leerob/directories; not a multi-model routing framework

CCPlugins (brennercruvinel) ★ 2.7k
ccplugins-cruvinel

24 professional Claude Code commands that prevent AI overengineering and automate repetitive development tasks.

10x-Tool-Calls ★ 863
10x-tool-calls

Maximize the number of tasks accomplished within a single Cursor/Windsurf request by running a user-input loop between tasks.

Clavix ★ 656
clavix

Enforces strict PRD/implementation phase separation via mandatory State Assertions and a 6-type Self-Correction Protocol across 20+ AI coding tools.

Claude Conductor (superbasicstudio fork) ★ 367
claude-conductor-superbasic

Gives Claude Code a persistent, cross-linked, auto-analyzed documentation system so it retains codebase context across sessions.

CCBot ★ 250
ccbot

Telegram bot that bridges to tmux-based Claude Code sessions, allowing remote monitoring and control via Telegram topics without interrupting the terminal sessi…

spec-based-claude-code ★ 129
spec-based-claude-code

Ten slash commands enforce requirements→design→tasks→implement with file-marker approval gates in plain Claude Code, zero dependencies.

opencode-plugin-openspec ★ 109
opcode-plugin-openspec

Prevent premature code implementation during OpenSpec planning by injecting a write-restricted planning agent into OpenCode.

AiDE ★ 106
aide

Standardizes AI context management across sessions with a zero-dependency markdown folder convention.

SDD (Solokit) ★ 61
sdd-session-ankushdixit

Eliminate AI context loss between sessions by reconstructing full work state from git history and spec files at each session start.

Aegis ★ 60
aegis

Gives AI assistants a shared project state machine so sessions don't lose context and tasks have verifiable state transitions.

agentic-agile-template (Microsoft) ★ 60
agentic-agile-template-ms

A GitHub template providing the CLAUDE.md, Issue templates, manifesto, and measurement framework teams need to adopt wave-based parallel human-agent development…

Personal OS (itseffi) ★ 33
agentic-os-effi

Provides a file-based personal operating system that grounds AI agent execution in explicit goals, backlogs, and 47 cross-domain skills spanning engineering, pr…

ai-ticket ★ 33
ai-ticket-jmike

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.

novel-writer-openspec ★ 29
novel-writer-openspec

Applies OpenSpec methodology to fiction writing to prevent character collapse and setting inconsistency in AI-assisted long-form novels.

Deliberate Agentic Development ★ 26
deliberate-agentic-dev

Structured AI agent workflow with Linear integration, human-review gates at every commit/PR/milestone, and state.json position tracking for precise resume.

spec-driver (davidlee) ★ 25
spec-driver-davidlee

Python CLI + 24 skills + TUI for evergreen specification management — specs that stay accurate as implementations evolve, not disposable planning documents.

AI Task Master (James-Cherished-Inc) ★ 21
ai-task-master-james

Fork of claude-task-master adapted for Roo Code, Cline, and OpenRouter multi-model API routing.

prd-creator-skill ★ 17
huchi996-prd-creator

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-r…

Arbor ★ 14
arbor-zed-gpui

Static API gateway framework (NOT an AI agent framework — slug mismatch in batch assignment).

oh-my-claude (ssenart) ★ 14
ssenart-oh-my-claude

Provides a real-time powerline status bar in Claude Code showing context usage, token count, Pro rate limit percentages, and reset timers.

alex-feel/claude-code-toolbox ★ 13
alex-feel-cc-toolbox

YAML-declarative meta-installer for Claude Code environments: define agents, MCP servers, hooks (command/HTTP/prompt/agent), model settings, and permissions in …

Claude-Slash-Commands (danielrosehill) ★ 10
claude-slash-commands-daniel

Provide a large library of reusable slash commands across 25 development domains for Claude Code.

wyndomb/claude-code-starter ★ 6
wyndomb-cc-starter

A productivity-workflow Claude Code starter for knowledge workers shipping daily planning, research, standup, and weekly synthesis commands with pre-wired Perpl…

AndAI ★ 5
andai

Self-hosted AI development loop using Redmine + Aider + YAML-configurable workflows with multi-model routing and human review gates.

OpenSpec Badge Action ★ 4
openspec-badge-action

Generate OpenSpec health metrics as SVG badges for embedding in GitHub READMEs via CI.

vibecodex ★ 4
vibecodex

54 numbered architecture rules (A1-F10) for FastAPI + Next.js + Go that AI coding agents follow automatically — enforcing layer boundaries, file-size limits, id…

OpenSpec.sh ★ 3
openspec-sh

POSIX bash reimplementation of OpenSpec for environments without Node.js — dependency-free spec-driven workflow scaffold.

SpecCrew ★ 3
speccrew

Apply ISA-95 industrial methodology to AI-assisted software delivery — 8 specialized agents orchestrate PRD through deployment with a 3-tier knowledge base and …

claude-commands (oddballteam) ★ 2
claude-commands-oddball

Provide analysis-only AI commands for engineering workflows with explicit no-automation policy for compliance-sensitive environments.

claude-overlay ★ 2
claude-overlay

Manages project-level Claude Code configuration for non-Anthropic model endpoints, surgically injecting/removing only overlay-managed keys while preserving all …

claude-snapshot ★ 2
claude-snapshot

Exports and restores the complete Claude Code configuration (~/.claude/) as a portable, diff-inspectable archive for multi-machine sync, OS reinstall recovery, …

OmniDev Kit ★ 1
omnidev-kit

Full-lifecycle AI development workflow skill for Claude Code: safety-first requirement alignment, adaptive phase planning, self-learning from session retrospect…

Agentic OS (KbWen) ★ 0
agentic-os-kbwen

Governance-first AI coding framework with classification-locked phase gates, SSoT state machine, and 26 commands + 14 skills that enforce No Evidence = No Compl…

CCCP / gendosu-claude-plugins ★ 0
cccp-gendosu

Provide a minimal plugin marketplace with TDD/git workflow commands and a meta-skill for creating new skills.

mcp-to-skill-converter ★ 0
mcp-to-skill-converter

Converts any MCP server into a Claude Code skill using progressive disclosure to achieve ~90% context savings at idle.

Specorator (Luis85/agentic-workflow) ★ 0
specorator-workflow

18 commands + 40 skills + 36 specialist agents + 11-stage lifecycle covering Idea through Retrospective — full software product lifecycle with dedicated orchest…

vitepress-plugin-openspec ★ 0
vitepress-plugin-openspec

Render an OpenSpec directory as structured VitePress documentation pages with auto-generated sidebar and nav.

acpx
acpx
Aegis (Folder Trees)
aegis-folder-trees

Community-referenced methodology using hierarchical folder trees as AI context and spec delivery mechanism.

bringyour-migration-auditor
asyncope-bringyour

unknown — repository returns HTTP 404

Axon
axon

unknown — repository not accessible

Blackbox Code
blackbox-code

unknown — repository not found (HTTP 404)

CCHub
cchub

CC ecosystem management platform with visual GUI (insufficient public material to verify).

Claude Oracle
claude-oracle

unknown — repository not accessible

claude-plugin-sdd (joestump)
claude-plugin-design-joestump
Flow Framework
flow-framework-khgs

unknown — repository not accessible

Forge (sudo-rebase)
forge-sudo-rebase

unknown — repository not accessible

GitAgentProtocol (OpenGAP)
gitagent-protocol
OmX (oh-my-codex-yeachan duplicate)
oh-my-codex-yeachan-omx

Duplicate entry — see canonical oh-my-codex-yeachan

OpenClaw
openclaw-os

AI agent runtime platform supporting plugin-based tool installation (osmoda-bridge), --local execution, multi-provider model routing (provider/model format), an…

sightglass
sightglass

unknown — repository not accessible

Slate V1
slate-v1

unknown — no public material available