Skill Claude CodeCodex
Strategic research framework that compresses months of market/competitive research into hours through structured power questions. Extracts unspoken industry insights, fragile market assumptions, and strategic attack surfaces from competitor data, reviews, and industry sources using parallel Exa-powered intelligence…
Skill Claude CodeCodex
ALWAYS invoke this skill when the user reports a bug to fix — it traces the active code path, writes a failing test, iterates the fix, and reviews the diff, with test gates that prevent fixing the wrong code path. Do not edit suspected code directly before tracing the active path and writing a failing test. Triggers…
Skill Claude CodeCodex
Conduct deep, multi-source research on any topic using parallel subagents, Exa search, and adversarial validation. Two modes: /deep-research (standard: 10+ sources, one researcher + a source-vetting pass) and /deep-research deep (multi-agent with 20+ sources, adversarial review, UNION merge). Use when the user asks to…
Skill Claude CodeCodex
Multi-pass first principles analysis for any domain — business, personal decisions, creative projects, career, health, relationships. Grounds the analysis in real evidence first, decomposes problems to fundamental truths, challenges assumptions through 4 universal lenses (Constraints, Resources, Human, Context), and…
Skill Claude CodeCodex
Spec-Driven TDD with a strict Red-Green-Refactor cycle using context-isolated subagents. Every feature starts with a structured spec; tests are generated from numbered acceptance criteria, with post-cycle three-dimension verification (Completeness + Traceability + Coherence), severity-tiered findings…
Agent Claude Code
Implement minimal code to pass failing tests for TDD GREEN phase. Reads the test file and writes minimum implementation. NEVER modifies test files. Raises specdefect: true instead of weakening tests when the spec is contradictory, unimplementable, or under-specified. Returns files, success output, and specdefect…
Agent Claude Code
Evaluate and refactor code after TDD GREEN phase. Improves quality while keeping tests green. Returns changes or "no refactoring needed" with reasoning.
Agent Claude Code
Write failing tests for TDD RED phase. Reads feature requirements and project documentation/specs to generate behavior-driven tests. Returns test file path and failure output. Enforces severity-aware assertion quality at source (rejects weak-only assertions, private-name leakage, excessive mocking).
Skill Claude CodeCodex
Semi-automatic end-of-session ritual. Scans the session and machine for loose ends, presents ONE categorized report, and on approval executes cleanup + closing tasks so the user can exit with nothing undone, uncommitted, unmerged, or left as garbage. Invoke whenever the user signals they're finishing up — "/wrap-up"…
Skill Claude CodeCodex
Deploy to staging/preview and prove it works before anything reaches users — runs pre-flight gates (types, tests, env, migrations, build), deploys, smoke-tests the live URL, diagnoses failures, and ends by writing a machine-readable verdict artifact that a release orchestrator can trust. Use for "/deploy-verify"…
Skill Claude Code
Enforce a filename convention for notes, docs, research, plans and transcripts at the moment a file is created, instead of documenting it and hoping. Ships a write-time PreToolUse hook that denies a non-conforming Write, a lint that is the single negative list both the hook and any periodic sweep call, and a…
Skill Claude Code
Universal production-release orchestrator — takes a FINISHED change to verified-released on any platform (web, bots, iOS, Android, automations) through a per-project recipe, verifying before AND after release and reporting honestly what "done" means for that platform (a store upload is never "live to users"). Crosses…
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: