190+ agent skills for Claude Code, Codex, Gemini CLI, Cursor, and 40+ other agents. Apache-2.0.
Agent Claude Code
Stage 2 of the /next-move meta-DAG. Breaks the Sensemaker's inferred problem into 3-7 subtasks with explicit dependencies and wave assignments. Pass the SensemakerOutput plus the ContextSnapshot. Returns DecomposerOutput JSON validated against schemas/decomposer-output.schema.json. Each subtask must be one skilled…
Agent Claude Code
Stage 3b of the /next-move meta-DAG. Identifies failure modes, severity, affected nodes, and concrete mitigations for the predicted plan. Pass the DecomposerOutput and the SkillSelectorOutput. Returns PreMortemOutput JSON validated against schemas/premortem-output.schema.json. Runs in parallel with the Skill Selector.…
Agent Claude Code
Stage 1 of the /next-move meta-DAG. Classifies a project's current state into well-structured / ill-structured / wicked, scores confidence, names the inferred problem, and decides whether to halt. Pass the gathered ContextSnapshot in your invocation. Returns SensemakerOutput JSON validated against…
Agent Claude Code
Stage 3a of the /next-move meta-DAG. For each subtask, picks a primary skill and a runner-up from the catalog using the SkillSearchService cascade (BM25, Tool2Vec, RRF, cross-encoder, attribution k-NN). Pass the DecomposerOutput plus a pre-narrowed candidate list per subtask. Returns SkillSelectorOutput JSON. Runs in…
Agent Claude Code
Stage 4 (final) of the /next-move meta-DAG. Assembles the final PredictedDAG from upstream outputs, picks planning + runtime topology with honesty rules applied, and validates against schemas/predicted-dag.schema.json. Pass the SensemakerOutput, DecomposerOutput, SkillSelectorOutput, and PreMortemOutput. Returns a…
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: