cryptofish7
25 verified skills
smoke-test
Generate and maintain a local deploy script (scripts/deploy.sh). Discovers project services, deploys them locally, and health-checks each one. Use when the user asks to "smoke test", "deploy locally", "test local deploy", "update deploy script", "run deploy", or "run smoke test".
testing
reflect
Analyze the current conversation for learnings and persist approved insights. Triggers on "reflect", "retrospective", "session review", "what did we learn", "conversation review".
tools
improve
Rewrite a vague prompt for clarity, then execute it. Use when the user invokes "/improve" or asks to sharpen, rewrite, or clarify their own prompt before acting on it.
documentation
codex
Review a plan file or a PR/branch diff using OpenAI Codex CLI. Use when the user says "/codex" (with or without an argument) or asks to review something with Codex. Auto-detects plan-review vs PR-review mode from the argument and conversation context.
tools
bug-bash-update
Update docs/BUG_BASH_GUIDE.md after implementing features or fixing bugs. Maintains a stable regression checklist organized by user-facing surface — folds changes into existing sections, annotates fixes, and verifies fixes in the browser. Triggers on "bug bash update", "update bug bash", "update testing checklist".
testing
refactor-audit
Audit a codebase for refactoring opportunities across all packages/sections. Spawns parallel refactor subagents per section, collects suggestions without executing, then presents a ranked plan for approval. Use when the user asks to "audit for refactors", "refactoring suggestions", "code quality review", "refactor plan", "find refactoring opportunities", or "what should we clean up".
development
skill-reviewer
Audit installed skills and agents for bloat, overlap, and improvement opportunities. Use when the user wants to review, audit, list, or improve their skills and agents. Triggers on "review skills", "audit skills", "list skills", "skill review", "improve skills".
testing
find-skills
Helps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express interest in extending capabilities. This skill should be used when the user is looking for functionality that might exist as an installable skill.
data-ai
smoke-test
Generate and maintain a local deploy script (scripts/deploy.sh). Discovers project services, deploys them locally, and health-checks each one. Use when the user asks to "smoke test", "deploy locally", "test local deploy", "update deploy script", "run deploy", or "run smoke test".
testing
setup-workflow
Set up the autonomous post-task workflow for a project. Injects the standard development pipeline into CLAUDE.md and installs all required skills and agents (docs-consolidator, ci-cd-pipeline, smoke-test, bug-bash-update, code-reviewer, debugger). Use at the start of a new project. Triggers on "setup workflow", "init workflow", "add workflow", or "set up project workflow".
development
ci-cd-pipeline
Analyze a repo and maintain its GitHub Actions CI/CD pipeline. Detects language, tooling, test frameworks, Docker, and deploy targets, then adds or removes workflow actions to match the project's current state. Use when the user asks to "add CI/CD", "update CI", "review pipeline", "set up GitHub Actions", "audit CI", "improve CI/CD", or any request about CI/CD pipelines, GitHub Actions workflows, or continuous integration.
tools
ci-cd-pipeline
Analyze a repo and maintain its GitHub Actions CI/CD pipeline. Detects language, tooling, test frameworks, Docker, and deploy targets, then adds or removes workflow actions to match the project's current state. Use when the user asks to "add CI/CD", "update CI", "review pipeline", "set up GitHub Actions", "audit CI", "improve CI/CD", or any request about CI/CD pipelines, GitHub Actions workflows, or continuous integration.
tools
codex-loop
Iteratively review and refine a plan file OR a PR/branch diff using OpenAI Codex CLI until no blocking issues remain. Use when the user says "/codex-loop" (with or without an argument) or asks to loop Codex review. Auto-detects plan-loop vs PR-loop mode from the argument and conversation context. Each round, Codex reviews as lead architect, Claude addresses blockers, then re-submits — until Codex signals ready or max iterations hit.
tools
dogfood
Systematically explore and test a web application to find bugs, UX issues, and other problems. Use when asked to "dogfood", "QA", "exploratory test", "find issues", "bug hunt", "test this app/site/platform", or review the quality of a web application. Produces a structured report with full reproduction evidence -- step-by-step screenshots, repro videos, and detailed repro steps for every issue -- so findings can be handed directly to the responsible teams.
development
quality-sweep
Full-codebase quality sweep via 8 parallel concern-focused subagents. Each owns one axis (dedup/DRY, shared types, dead code, circular deps, weak types, defensive programming, legacy/fallbacks, AI slop/comments), researches, produces a critical assessment, and implements high-confidence fixes. Use when the user asks to "clean up the codebase", "quality sweep", "vibecode cleanup", "remove slop", "kill dead code and any types", or wants a comprehensive cross-cutting cleanup rather than a per-package refactor audit.
development
quality-sweep
Full-codebase quality sweep across 8 concern-focused lanes: deduplication/DRY, shared types, dead code, circular dependencies, weak types, defensive programming, legacy fallbacks, and AI slop/comment hygiene. Use when the user asks to "quality sweep", "clean up the codebase", "remove slop", "kill dead code", "fix weak types everywhere", or wants a broad cross-cutting cleanup executed rather than a per-package refactor audit.
development
ralph
Autonomously work through all project tasks. Finds the next incomplete task, implements it, runs the full post-task pipeline (verify, docs, CI/CD, commit, PR, review, CI, merge), then moves to the next task. Keeps going until all tasks are done. Triggers on "ralph", "work on tasks", "run tasks", "autopilot", or "do the work".
development
ralph
Autonomously work through all project tasks. Finds the next incomplete task, implements it, runs the full post-task pipeline (verify, docs, CI/CD, commit, PR, review, CI, merge), then moves to the next task. Keeps going until all tasks are done. Triggers on "ralph", "work on tasks", "run tasks", "autopilot", or "do the work".
development
reflect
Analyze the current conversation for learnings and persist approved insights. Triggers on "reflect", "retrospective", "session review", "what did we learn", "conversation review".
tools
refactor-audit
Audit a codebase for refactoring opportunities across all packages/sections. Spawns parallel refactor subagents per section, collects suggestions without executing, then presents a ranked plan for approval. Use when the user asks to "audit for refactors", "refactoring suggestions", "code quality review", "refactor plan", "find refactoring opportunities", or "what should we clean up".
development
skill-reviewer
Audit installed skills and agents for bloat, overlap, and improvement opportunities. Use when the user wants to review, audit, list, or improve their skills and agents. Triggers on "review skills", "audit skills", "list skills", "skill review", "improve skills".
testing
docs-consolidator
Audit and consolidate project documentation in the docs/ folder, including CLAUDE.md optimization. Use when the user wants to clean up docs, check docs are up to date, deduplicate information across docs, ensure information lives in the right doc, reorganize documentation, or slim down CLAUDE.md files. Triggers on "consolidate docs", "clean up documentation", "audit docs", "organize docs", "sync docs with code", "audit claudemd", "review claude.md", "slim down claude.md", "optimize claude.md", "claudemd audit".
development
bug-bash-update
Update docs/BUG_BASH_GUIDE.md after implementing features or fixing bugs. Maintains a stable regression checklist organized by user-facing surface — folds changes into existing sections, annotates fixes, and verifies fixes in the browser. Triggers on "bug bash update", "update bug bash", "update testing checklist".
testing
docs-consolidator
Audit and consolidate project documentation in the docs/ folder, including AGENTS.md optimization and CLAUDE.md migration. Use when the user wants to clean up docs, check docs are up to date, deduplicate information across docs, ensure information lives in the right doc, reorganize documentation, or slim down AGENTS.md / CLAUDE.md files. Triggers on "consolidate docs", "clean up documentation", "audit docs", "organize docs", "sync docs with code", "audit agents.md", "review agents.md", "audit claude.md", "review claude.md", "slim down agents.md", "optimize agents.md", "slim down claude.md", "optimize claude.md".
development
setup-workflow
Set up the autonomous post-task workflow for a project. Injects the standard development pipeline into AGENTS.md and installs all required skills and agents (docs-consolidator, ci-cd-pipeline, smoke-test, bug-bash-update, code-reviewer, debugger). Use at the start of a new project. Triggers on "setup workflow", "init workflow", "add workflow", or "set up project workflow".
development