Skill Claude CodeCodex
Multi-perspective code review with simpler-alternative pass. When NOT to use: deep adversarial analysis (use council); simple code review without quality concerns.
Skill Claude CodeCodex
Metrics, diagnostics, correlation, retry, deadletter, and recovery evidence workflow. When NOT to use: one-off event capture (use event-log-tracing); pure recovery logic (use async-worker-recovery).
Skill Claude CodeCodex
Multi-phase orchestration for planners and executors. When NOT to use: single-task edits (use delegation-patterns); pure read-only audits (use read-only-explorer); one-shot questions (use ask).
Skill Claude CodeCodex
Session ownership and authorization workflow. When NOT to use: general session lifecycle questions (use live-agent-lifecycle); for non-session auth issues.
Skill Claude CodeCodex
Pi extension lifecycle and registration patterns. When NOT to use: single registration event (use direct API calls); for child worker lifecycle (use child-pi-spawning).
Skill Claude CodeCodex
Write engineering RCA record after bug is fixed. When NOT to use: unresolved bugs (use systematic-debugging first); minor fixes that don't warrant RCA.
Skill Claude CodeCodex
Read-only exploration and audit workflow. When NOT to use: any task that needs write actions; deep adversarial analysis (use council); quick one-line lookups (use bash directly).
Skill Claude CodeCodex
Use when a goal, issue, roadmap item, review finding, or user request must become actionable worker tasks. When NOT to use: already-decided work that just needs execution; tasks so simple they fit one sentence.
Skill Claude CodeCodex
Deep-research skill combining iterative depth, structured+validated output, rigor mechanisms, anti-thrash, and pi-native hooks for general deep research. REQUIRED — read the full skill file first (iterative-depth protocol with rigor scripts); run verifycitations and sourceevaluator on your output before claiming done.…
Skill Claude CodeCodex
Use this skill for pi-crew resource/config work.
Skill Claude CodeCodex
Safe read-only navigation of pi-crew run state. When NOT to use: write actions or mutations (use state-mutation-locking); configuration changes (use resource-discovery-config).
Skill Claude CodeCodex
Safe shell-command workflow. When NOT to use: subprocess orchestration (use child-pi-spawning); destructive operations without sandbox (use ownership-session-security).
Skill Claude CodeCodex
Outsider-perspective review questioning intent before tracing code. When NOT to use: implementation work (use systematic-debugging); trivial one-line fixes that don't warrant outsider review.
Skill Claude CodeCodex
Use when reviewing delegation, skill loading, tool access, worker prompts, artifacts, runtime config, state, ownership, or subprocess execution. When NOT to use: pure code quality issues without security implications (use multi-perspective-review); single-file fixes (use scrutinize).
Skill Claude CodeCodex
Durable state mutation and locking workflow. When NOT to use: read-only inspection (use runtime-state-reader); single-process non-concurrent edits.
Skill Claude CodeCodex
Four-phase debugging discipline with refuse gates. When NOT to use: trivial fixes where root cause is obvious; production incidents needing immediate rollback (use post-mortem after).
Skill Claude CodeCodex
Evidence before claims. When NOT to use: mid-task self-correction (use scrutinize); pure execution tasks where output IS evidence.
Skill Claude CodeCodex
Pi TUI crew widget data sources, display priority, and rendering performance. When NOT to use: non-TUI displays; backend data sources without widget context (use runtime-state-reader).
Skill Claude CodeCodex
Workspace isolation boundaries. When NOT to use: concurrent edits within same repo (use worktree-isolation); non-git projects.
Skill Claude CodeCodex
Conflict-safe git worktree workflow. When NOT to use: non-git changes (use direct edit); single-commit patches (use git-master).
Agent Claude Code
Critic seat on a decision council. Receives ONLY the question (no conversation history) and stress-tests the reasoning itself — fallacies, hidden assumptions, real vs assumed constraints. Spawned by the council skill. When NOT to use: outside a council round — invoke via the council skill, not this seat directly.
Agent Claude Code
Pragmatist seat on a decision council. Receives ONLY the question (no conversation history) and weighs practical tradeoffs — cost, maintenance, time-to-value, alternatives. Spawned by the council skill. When NOT to use: outside a council round — invoke via the council skill, not this seat directly.
Agent Claude Code
Skeptic seat on a decision council. Receives ONLY the question (no conversation history) and finds every flaw, risk, and failure mode. Spawned by the council skill. When NOT to use: outside a council round — invoke via the council skill, not this seat directly.
Agent Claude Code
UI/UX specialist for designing, reviewing, and implementing interfaces — web frontends and terminal UIs. Use for styling, layout, interaction, and design handoff. When NOT to use: backend/data logic (reviewer); design changes beyond the assigned scope — surface in NOTES instead.
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: