Skip to content
/

Primitives

45 one-of-a-kind primitives

Architectural primitives that appeared in exactly one (or two tightly-related) frameworks across the entire 654-framework corpus. Extracted verbatim from the Phase E executive synthesis. Each represents a genuinely novel engineering decision — not a variation on an existing pattern.

Source: _index/MASTER-BY-DIMENSION-v2.md §19

#1
Memory

HTTP-proxy LLM-call interception with hallucination-detection certificates

entroly runs on :9377, compresses context via knapsack DP (Rust), attaches WITNESS NLI certificates (AUROC 0.80 on HaluEval-QA). The only memory-layer framework that sits below the agent rather than beside it.

entroly single-source
#2
Hooks

Auto-backpropagation of runtime failures into the spec

PostToolUse hook detects runtime errors and adds R-numbered acceptance criteria + regression tests back into the originating spec. Living spec, not authored spec.

forge-lucasduys single-source
#3
Hooks

LLM-evaluated hooks ("type":"prompt")

First surfaced in cc-spec-driven, devteam-michael-harris. Phase D revealed 5+ independent implementations — now its own archetype A21.

cc-spec-driven single-source
#4
Governance

Sentinel-driven state machine in stdout

AI emits ✅ Ready / ⛔ Blocked markers; PostToolUse hooks parse stdout into review-gate state.

sd0x-dev-flow single-source
#5
Governance

Ralph-loop XML completion contract

Machine-detectable completion signal embedded in natural-language output: <promise>TASK_COMPLETE</promise>. Independently implemented in ralphy-openspec, cursor-rules-pack-v2.

ralphy-openspec single-source
#6
Governance

Adversarial duel review

Two agents independently review the same implementation, then cross-score each other. The only formal adversarial-review protocols in the corpus.

ag-coding-flywheel single-source
#7
Governance

Fuzzy Linear Temporal Logic monitor

Formal LTL ordering contracts at 0.139 ms p50.

sponsio single-source
#8
Hooks

Parallel LLM-powered execution sentinels

Asynchronous LLM observers tap the event stream, detect stuck states, trigger recovery.

hankweave single-source
#9
Compression

"Think in Code" — agent writes analysis scripts instead of reading data into context

100× context savings vs any other technique.

context-mode single-source
#10
Memory

JIT semantic context loader via MCP resources

1,675 useWhen scenarios + 4,036 keywords + LRU cache deliver 95–98% token reduction.

orchestr8 single-source
#11
Governance NEW D

TESTABILITY pre-implementation Opus veto

A test-designer Opus subagent returns PASS/WARN/FAIL on the Plan agent's class design before any code is generated. Only example in the corpus of structured design-time veto between agents.

#12
Hooks NEW D

Stop-hook drift-log compliance gate (LLM-on-LLM at Stop)

The Stop hook itself ships session context to another LLM that evaluates whether the agent obeyed its drift log; if not, it blocks completion. The most aggressive use of LLM-evaluated hooks.

atmosphere single-source
#13
Governance NEW D

Compiled Go binary "don't self-approve" code invariant

The "don't mark your own work done" rule is a Go compile-time check rather than a prompt, removing the possibility of the agent talking itself around it.

liza single-source
#14
Hooks NEW D

PreToolUse banner injection as orchestrator→subagent message bus

Hooks (PreToolUse, UserPromptSubmit, SessionStart) inject [EVO DIRECTIVE] banners that act as a downward communication channel to in-flight subagents. The only framework treating hooks as IPC.

evo single-source
#15
Self-improvement NEW D

Karpathy autoresearch loop (empirical SKILL.md optimization)

The framework writes PRs against itself to optimize its own skill prompts empirically. Crosses Phase C's ag-coding-flywheel adversarial-review into closed-loop self-modification.

#16
Governance NEW D

Compiled YAML policy → PreToolUse hooks (5-level escalation)

Policy DSL compiles into a hierarchy of PreToolUse hooks with progressive enforcement: warn → block → quarantine → revert → escalate. The only structural escalation pattern in the corpus.

dotforge single-source
#17
Governance NEW D

Loop-detection BREAK-LOOP PROTOCOL

When Claude repeats the same (phase, nextAction) 3+ times, the framework injects a numbered escape menu: try alternative / escalate / upgrade strategy. The only meta-cognitive self-recovery primitive in the corpus.

ccg-workflow single-source
#18
Governance NEW D

8-dimension partnership measurement + wave-based file ownership

Quantitative agent-human partnership scoring across 8 dimensions, with file ownership rotated across waves so no single contributor (human or AI) dominates a region.

#19
Substrate NEW D

NixOS atomic rollback + AI root access + post-quantum mesh

Give the AI full root, but make every commit a NixOS generation that can be rolled back atomically. Only OS-distro framework in the corpus.

osmoda single-source
#20
Hooks NEW D

PreToolUse-on-Task hook for sub-agent context injection

When the parent agent calls Task to spawn a sub-agent, the PreToolUse hook injects curated JSONL spec files into the sub-agent's prompt. "Specs injected, not remembered."

trellis single-source
#21
Self-improvement NEW D

Self-evolving plugin meta-linter (auto-PRs new rules)

Detects new failure patterns and auto-PRs new lint rules into the framework itself. The closest the corpus comes to a self-modifying agent runtime.

nlpm-xiaolai single-source
#22
Governance NEW D

Hard USD threshold enforcement

Actually stops execution at a configurable USD cap rather than warning. Only example in the corpus.

octomind-muvon single-source
#23
Substrate NEW D

Compose-lang: framework that only defines a new language

Ships a DSL spec and a parser, no agents, no skills. Meta-level "what if the framework is a syntax?" answer.

compose-lang single-source
#24
Routing NEW D

Only true peer-to-peer swarm

Agents handoff to each other without a central coordinator; swarm pattern in literal sense. Phase C's claude-flow was queen+workers, not P2P.

mcp-agent-lastmile single-source
#25
Substrate NEW D

Claude-Code-as-containerized-executor inversion

A Go binary embeds V8 and containerizes Claude Code itself as a capability being orchestrated rather than the orchestrator. Inverts the normal relationship.

yao single-source
#26
Compression NEW D

~75% token reduction via symbolic encoding + 6-step backprop

Caveman notation §G/§C/§I/§V/§T/§B compresses spec text by ~75%; every test failure backprops into a §B entry and §V invariant.

#27
Routing NEW D

Sonnet-classifier-at-UserPromptSubmit shaping response mode/tier

The UserPromptSubmit hook fires a Sonnet classifier on the user prompt itself and selects which response tier/persona Claude operates under. Mode-aware agent.

#28
Compression NEW D

88% token efficiency replacing MCP tool-calling with code-exec

Instead of MCP tool calls, the agent generates code that calls a library. 88% fewer tokens than equivalent MCP usage.

code-mode-library single-source
#29
Hooks NEW D

LLM-evaluated Stop hook

The Stop hook ships full session context to an LLM that returns block/continue.

cc-spec-driven single-source
#30
Governance NEW D

Bug Council 5-analyst parallel escalation

"type":"prompt" hooks with 5-analyst parallel escalation and SQLite per-task model history.

#31
Governance NEW D

ISA-95 industrial-controls methodology applied to AI workflows

Borrows enterprise integration standard from manufacturing.

speccrew single-source
#32
Memory NEW D

27-tool MCP self-awareness server

The MCP server's tools are tools to inspect the running MCP server itself. Meta-introspection.

hawkeye single-source
#33
Substrate NEW D

50+ module-scoped CLAUDE.md files

Pushes Claude Code's CLAUDE.md hierarchy further than any other framework — one per module, scoped by directory.

ptah-orchestra single-source
#34
Governance NEW D

LSP linter with 423 validation rules

The only LSP server in the corpus dedicated to linting agent context files (CLAUDE.md, AGENTS.md).

agnix single-source
#35
Substrate NEW D

AGENTS.md ecosystem-win over AGENT.md (270× stars)

Demonstrates that naming convention adoption can win despite technical inferiority. The community converged on AGENTS.md (plural) even though AGENT.md (singular) had slightly better technical design.

agents-md single-source
#36
Governance NEW D

state:agent-ready self-application prohibition

The agent cannot set its own ready state; only an external signal can.

openshell-nvidia single-source
#37
Routing NEW D

Cross-vendor model routing (Claude→GPT-5/Qwen)

7+ frameworks now route across competing vendors; A16 archetype. ccg-workflow, droids-curtis, switchboard, octomind-muvon, golembot.

ccg-workflow single-source
#38
Governance NEW D

Auto-merge-on-CI-pass

If CI passes, the PR auto-merges. The only framework with merges_automatically: yes plus CI gating.

cueapi single-source
#39
Self-improvement NEW D

Self-hosting / self-developing

The framework is used to develop the framework itself, dogfooding the full SDLC loop.

specorator single-source
#40
Governance NEW D

policy-as-code DSL (safe-outputs separation)

Agentic workflows as policy: every write goes through a separately sandboxed job; safe-outputs separation. The most security-conscious framework in the corpus.

gh-aw-githubnext single-source
#41
Routing NEW D

Tap registry pattern (community specialists as one-file TOML manifests)

octomind tap <github-repo> installs a community agent like a Homebrew tap.

octomind-muvon single-source
#42
Memory NEW D

Epistemic lease / decay model

Cross-module trajectory knowledge decays over time and must be renewed via 4 escalation levels.

spec-gen single-source
#43
Routing NEW D

13-tier cost-based model routing

T0 local free → T12 Opus $25/M, with iCPG and Telos IFS = F1×F2×F3 quality scoring. Most elaborate multi-model config in the corpus.

claude-bootstrap single-source
#44
Compression NEW D

rtk-token-killer (54k★ beating every orchestration framework)

Single-purpose token-efficiency tool. Proves the community values cost optimization over orchestration.

rtk-token-killer single-source
#45
Routing NEW D

Cross-vendor routing as visual kanban

Routes Gemini Flash for cheap tasks, Claude Opus for expensive ones, via a VS Code kanban webview.

switchboard single-source