kaijutsu

Diagnostic + cleanup tool for AI agent state directories. Audits ~/.claude, ~/.codex, ~/.gemini, ~/.agents; surfaces stale sessions, orphaned cache, broken hook refs; reclaims space safely (trash-not-delete, 30-day recovery).

Third-party skill from github.com/addyosmani/agent-skills; install via `jutsu install api-and-interface-design`.

Multi-pass adversarial critique. Run a skeptic review N times in fresh contexts (optionally in parallel via dispatch-parallel); deduplicate; surface only issues that survive.

Third-party skill from github.com/addyosmani/agent-skills; install via `jutsu install browser-testing-with-devtools`.

Third-party skill from github.com/addyosmani/agent-skills; install via `jutsu install ci-cd-and-automation`.

Third-party skill from github.com/addyosmani/agent-skills; install via `jutsu install code-review-and-quality`.

Reduce code without changing behavior. Apply Chesterton's Fence + Rule of 500. Find dead code, duplicate logic, unused abstractions, leaky helpers, premature factoring. Composes polish (review-fix loop) and blunder-hunt (regression risk).

Third-party skill from github.com/addyosmani/agent-skills; install via `jutsu install context-engineering`.

Detect when an iterative loop has stopped producing new signal. Quantitative scoring (token deltas, jaccard similarity, new-item ratio) plus optional persistence of round-by-round signals to .claude/convergence-log/ for post-hoc tuning.

Destructive command guard. Pre-tool-use hook that blocks rm -rf /, git reset --hard, git clean -fd, rm of .env / credentials, and other unrecoverable shell commands before the agent executes them.

Third-party skill from github.com/addyosmani/agent-skills; install via `jutsu install debugging-and-error-recovery`.

Decision journal — record architectural decisions with reasoning + tripwires; multi-model critique on record; auto-tag from content; surface recorded decisions when about to contradict one. Writes to project-memory.

Third-party skill from github.com/addyosmani/agent-skills; install via `jutsu install deprecation-and-migration`.

Strip AI-generated tells from a draft. Two modes: --strip edits in place; --detect reports flagged passages without editing. Reads project style guides from CLAUDE.md/AGENTS.md for project-specific term preservation.

Fan out N independent tasks to parallel subagents, collect, synthesize. Primitive for skills that need parallelism (multi-lens review, multi-file edits, multi-source data gathering).

Third-party skill from github.com/addyosmani/agent-skills; install via `jutsu install documentation-and-adrs`.

Third-party skill from github.com/addyosmani/agent-skills; install via `jutsu install frontend-ui-engineering`.

Third-party skill from github.com/addyosmani/agent-skills; install via `jutsu install git-workflow-and-versioning`.

Third-party skill from github.com/addyosmani/agent-skills; install via `jutsu install idea-refine`.

Implement features as thin vertical slices, each shippable behind a feature flag. Each slice ends in green CI. No big-bang refactors. Composes scope-check (cost-tier) and polish (per-slice review).

Periodic development journal — read git/PRs/decisions/memory; produce a structured narrative; optional multi-model synthesis for the narrative voice. Themed cadences (weekly/sprint/release/monthly) and sentiment+velocity tracking.

Prompt primitive — assert N issues exist to force exhaustive enumeration. Includes a transparency footer recommendation: every output that used lie-to-them should disclose the claimed-N and actual-found counts.

Fan out the same task to multiple models (Claude, Codex, Gemini), compare responses, synthesize a 'best of all worlds' answer that artfully blends each model's strengths.

Third-party skill from github.com/addyosmani/agent-skills; install via `jutsu install performance-optimization`.

Decompose a spec into a verifiable task list — vertical slices, explicit dependencies, acceptance criteria per task. Final review pass via blunder-hunt + scope-check before tasks are considered ready for implementation.

Automated post-implementation review-and-fix loop. Up to 4 passes; each pass alternates random-code-exploration and cross-agent-review lenses; convergence-detect declares the stop; deslop runs at the end on any user-facing prose.

Adversarial PR review. Run by a different agent than the committer; assume bugs exist; surface them as inline PR comments via gh. Optional multi-model fan-out for high-stakes diffs.

Cross-skill project-memory contract. Defines the schema and conventions for the per-project memory directory that decide, session-retro, unstuck, and other skills read and write.

Detect README staleness, identify out-of-date sections, propose targeted diffs. Semantic section matching (Quickstart ≈ Get Started); works on any markdown doc (CONTRIBUTING, ARCHITECTURE, CHANGELOG); auto-generates usage examples from new CLI commands; checks badge staleness.

Mid-task scope mirror with cost-tier classification (plan/bead/code = 1x/5x/25x). Surfaces when work has drifted from cheap planning into expensive code-space rework, and identifies natural split-points where a PR could become two.

OWASP-Top-10-aware security review. Sweep the diff (or a target file set) for injection, auth gaps, deserialization risks, broken access control, sensitive data exposure, security misconfig, and supply-chain trust. Composes blunder-hunt's hostile-input lens.

End-of-session retrospective. Mines the current conversation for learnings; cross-session pattern detection across recent retros; writes to project-memory; optional auto-apply mode.

Third-party skill from github.com/addyosmani/agent-skills; install via `jutsu install shipping-and-launch`.

Third-party skill from github.com/addyosmani/agent-skills; install via `jutsu install source-driven-development`.

Author a PRD-style spec before writing implementation. Produce a written, reviewable artifact (problem statement, scope, non-goals, acceptance criteria, open questions) that survives the agent session and gets reviewed by a fresh-eyes pass before unblocking implementation.

Third-party skill from github.com/addyosmani/agent-skills; install via `jutsu install test-driven-development`.

Guided problem articulation when you're stuck — break unproductive loops by forcing clarity, then escalate via blunder-hunt + structured pivots.