Skill Claude CodeCodexCursor
Audits architecture, finds duplicates, analyzes dependencies. Use when TRIGGERAUDITOR fires (every 200 tasks) or when architecture drift is suspected.
not rated 1 5mo ago A 37 tokens
original MIT
1Stars on the repository
27Mods indexed here, across every type
5mo agoLast push, which is what freshness is scored on
MITLicence, which decides whether bodies are shown
Skill Claude CodeCodexCursor
Executes micro-tasks: writes code, runs tests, fixes errors, updates task status. Use when the scheduler assigns a task or when implementing a specific task from the task list.
not rated 1 5mo ago A 41 tokens
original MIT
Skill Claude CodeCodexCursor
Analyzes technical debt, optimizes DAG, resplits tasks. Use when TRIGGERMETAPLANNER fires (every 50 tasks) or when DAG needs optimization.
not rated 1 5mo ago A 39 tokens
original MIT
Skill Claude CodeCodexCursor
Expands one epic into 15–25 subtasks. ONLY Planner uses this. Left/Right NEVER call planner-sub — they call planner.
not rated 1 5mo ago A 33 tokens
original MIT
Skill Claude CodeCodexCursor
Planner breaks goals into epics and dispatches Sub-Planner per epic. Planner does NOT create task files — Sub-Planner does.
not rated 1 5mo ago A 32 tokens
original MIT
Skill Claude CodeCodexCursor
Summarizes project state, updates architecture, compresses context. Use when TRIGGERRESEARCHER fires (every 20 tasks) or when context has grown too large.
not rated 1 5mo ago A 41 tokens
original MIT
Skill Claude CodeCodexCursor
Reviews code for quality, correctness, and adherence to architecture. Use after each task completion or when the user requests a code review.
not rated 1 5mo ago A 31 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: