Agent Claude Code
Drafts and revises implementation plans. Used by the make-detail-plan skill in a planner/reviewer discussion loop.
not rated 3
changed 4d ago A 27 tokens
original MIT
3Stars on the repository
73Mods indexed here, across every type
todayLast push, which is what freshness is scored on
MITLicence, which decides whether bodies are shown
Agent Claude Code
CC fallback implementation-plan reviewer for make-detail-plan; invoked when codex CLI is unusable. Thorough — surfaces minor issues as well as major ones.
not rated 3
changed 4d ago A 34 tokens
original MIT
Agent Claude Code
3-pass GitHub issue dedupe survey. Classifies verdict (none/reopen/sub-of/make-parent/sibling) and writes a JSON artifact. Read-only — no issue creation, no comments.
not rated 3
changed 9d ago B 46 tokens
original MIT
Agent Claude Code
Proposes 2-3 mutually-exclusive high-level approaches for a task. Used by the make-outline-plan skill. Inspired by Aider's architect/editor split and GitHub Spec Kit's /specify stage.
not rated 3
changed 4d ago A 46 tokens
original MIT
Agent Claude Code
CC fallback approach reviewer for make-outline-plan; invoked when codex CLI is unusable. Checks direction and coverage only — never implementation details.
not rated 3
changed 4d ago A 32 tokens
original MIT
Agent Claude Code
CC fallback security-plan reviewer for review-plan-security; invoked when codex CLI is unusable.
not rated 3
changed 21d ago A 24 tokens
original MIT
Agent Claude Code
LLM judge for /refactor-prompts. Classifies hot regions from the lexical scan and emits an edit plan JSON.
not rated 3 today A 32 tokens
original MIT
Agent Claude Code
Scan code for security anti-patterns across three axes. Read-only, no web access. Used by review-code-security.
not rated 3
changed 9d ago A 28 tokens
original MIT
Agent Claude Code
EM Supervisor — audit mode review agent. Invoked by Stop-hook block at stage boundaries (CONFIRMINTENT/OUTLINE/DETAIL) or when cumulative severity reaches the audit threshold. Reviews cross-stage coherence and writes a verdict (CONTINUE/WARN/BLOCK) to the supervisor state file.
not rated 3
changed 3d ago A 64 tokens
original MIT
supervisor
10Agent Claude Code
EM Supervisor — alert mode review agent. Invoked by Stop-hook block when a sentinel-hang, scheduled-review, or off-proposal condition is detected. Reviews the active session against JD checklist and writes findings to the supervisor state file.
not rated 3
changed 3d ago A 50 tokens
original MIT
survey-code
11Agent Claude Code
Investigate the codebase to understand existing patterns, constraints, and relevant files. Writes a session-scoped survey-code.md artifact.
not rated 3
changed 1mo ago A 29 tokens
original MIT
Agent Claude Code
Investigate git history and GitHub issue/PR timeline since the relevant issue opened, to surface changes that may invalidate the issue's premises.
not rated 3 today A 31 tokens
original MIT
Agent Claude Code
CC fallback test-coverage reviewer for review-tests; invoked when codex CLI is unusable.
not rated 3
changed 4d ago A 23 tokens
original MIT
Agent Claude Code
Research an external topic via WebSearch and WebFetch. Read-only. Used by deep-research.
not rated 3
changed 9d ago A 25 tokens
original MIT
Agent Claude Code
Dedicated opus-fixed subagent for judging task complexity signals. Called by skills when no persisted evaluation exists. Returns exactly one SIGNALS: line.
not rated 3
changed 3d ago A 0 tokens
original MIT
Agent Claude Code
Dedicated sonnet subagent for the outline-skip and detail-skip judgments. Called by jev-judge when Jev cannot decide every key. Returns exactly one SKIPCONDITIONS: line.
not rated 3
changed 3d ago A 0 tokens
original MIT
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: