Skill Claude CodeCodex
Explain a file, function, or concept with layered depth — from high-level overview to implementation details. Adapts to the reader's experience level. Triggers: "explain", "how does this work", "what does this do", "walk me through". Do NOT use this skill for: generating documentation (use a writer), debugging (use…
Skill Claude CodeCodex
Clean up stale git branches, prune remotes, and tidy repository state. Removes merged branches and shows orphaned work. Triggers: "git cleanup", "clean branches", "prune branches", "tidy git". Do NOT use this skill for: resolving merge conflicts, rebasing, or any destructive operations on unmerged work.
Skill Claude CodeCodex
Write a structured session handoff summary so the next session picks up cleanly. Captures what was done, what's left, decisions made, and gotchas. Triggers: "handoff", "session summary", "wrap up", "save progress", "end of session". Do NOT use this skill for: mid-session notes, creating plans for new work, or…
Skill Claude CodeCodex
Use when the user wants to design, redesign, shape, critique, audit, polish, clarify, distill, harden, optimize, adapt, animate, colorize, extract, or otherwise improve a frontend interface. Covers websites, landing pages, dashboards, product UI, app shells, components, forms, settings, onboarding, and empty states.…
Skill Claude CodeCodex
Download a Loom video share URL and produce an audio transcript (whisper) plus optional keyframes for visual analysis. Triggers on any loom.com share URL the user provides, or on phrases "transcribe this loom", "analyze this loom", "watch this loom", or "/loom-analyze" followed by a URL.
Skill Claude CodeCodex
Safe database migration workflow with backup verification, dry-run, and rollback plan. Triggers: "migrate database", "run migration", "db migration", "schema change". Do NOT use this skill for: writing queries, debugging query performance, or designing schemas from scratch. Only for executing schema migrations safely.
Skill Claude CodeCodex
Cross-terminal morning briefing. Consolidates every parked Claude Code session into one report — branch, uncommitted state, recent asks, last action — so you resume 6-10 terminals without rereading each. Triggers: "morning", "morning briefing", "what were my terminals doing", "where did I leave off", "/morning". Do…
Skill Claude CodeCodex
Cross-check answer across 5 AI CLIs (claude/codex/gemini/copilot/kiro) in parallel + adversarial review. Auto-trigger only for high-stakes: security-critical code, prod incidents, scale architecture (over 1k RPS), irreversible decisions (DB migrations, breaking API changes). Or explicit /multiask. Cost 1M tokens +…
Skill Claude CodeCodex
Generate a MERGED, all-logins Claude Code insights report — the /insights equivalent that covers your ENTIRE local session corpus (every account you've logged in as on this machine), not the built-in's small sample. Quantitative rollups from all session-meta + a qualitative narrative synthesized from the local facets…
Skill Claude CodeCodex
Performance investigation workflow: profile, identify bottlenecks, and optimize with measurable before/after results. Prevents premature optimization. Triggers: "perf check", "performance", "slow", "optimize", "bottleneck". Do NOT use this skill for: general code quality (use /code-review), fixing bugs (use /debug)…
Skill Claude CodeCodex
Batch review and manage multiple open PRs in one pass. Lists all open PRs, checks CI status, reviews diffs, and provides a consolidated report with merge recommendations. Triggers: "review all PRs", "batch PR review", "check my PRs", "PR status".
Skill Claude CodeCodex
Parallel PR processing via subagent fleet + git worktrees, autonomous overnight (default) or interactive batched with user checkpoints (--interactive flag). Coordinator triages backlog, spawns one worker per PR (capped at 3 concurrent), only escalates protected-path or non-trivial conflicts to user. Triggers…
Skill Claude CodeCodex
Distill the current task and state into a clean, self-contained reprompt to paste into a fresh session, so a stale or bloated context can be cleared without losing the thread. Triggers "/reboot", "reboot this session", "this session is stale", "stale session", "clear and reprompt", "reprompt me", "fresh start", "this…
Skill Claude CodeCodex
Focused refactoring with zero behavior change guarantee. Applies rename, extract, inline, or simplify operations with mandatory test verification before and after. Triggers: "refactor", "clean up", "extract function", "rename", "simplify". Do NOT use this skill for: adding features, fixing bugs, or any change that…
Skill Claude CodeCodex
Enforce strict research-only / analysis-only mode. When invoked, Claude MUST NOT use Edit, Write, MultiEdit, or NotebookEdit tools — findings are delivered as markdown only. Use when the user says "research-only", "analysis-only", "don't change code, just investigate", or at the start of a BMAD/ralph research phase.…
Skill Claude CodeCodex
Persistent low-effort output mode for tired days. Make every reply dead-simple to read — a bold one-line answer, then one next step, nothing else unless asked. Plain everyday words, no jargon, no options menus, no preamble. Stays on across turns until turned off. Triggers "/rest", "rest mode", "tired mode", "im…
Skill Claude CodeCodex
Resume and re-run the last task that failed because of a Claude API error or outage (overloaded/529, 500/503, rate limit, network drop, "Claude is down"). Use AFTER the error clears, when the user says "/retry", "retry", "try that again", "it errored, redo it", "pick up where you crashed", or "the API was down…
Skill Claude CodeCodex
Quick security scan of recent changes. Checks for OWASP Top 10 patterns including hardcoded secrets, injection vulnerabilities, XSS, and insecure dependencies. Triggers: "security check", "security scan", "check for vulnerabilities", "audit security". Do NOT use this skill for: full penetration testing, compliance…
Skill Claude CodeCodex
This skill is awaiting contribution. See references/ for framework-specific implementation files.
Skill Claude CodeCodex
This skill is planned. It will take scanner findings (e.g. Checkmarx CSV), load the relevant OWASP cheat sheet as context, apply the fix, and verify with a re-scan.
Skill Claude CodeCodex
End-of-session doubt enumeration before claiming substantive work done. Surfaces (1) what the agent is least confident about — then root-causes each item — and (2) the biggest thing the user may not realize about the situation. A two-question reflective close-out, distinct from test-running verification. Triggers…
Skill Claude CodeCodex
Best practices for building Claude Code / Claude.ai Skills, distilled from Anthropic's official "Complete Guide to Building Skills". Use when creating a new skill, writing or fixing a SKILL.md, choosing a folder name or description, debugging why a skill over- or under-triggers, or reviewing a skill before shipping.…
Skill Claude CodeCodex
Capture learnings as reusable skills, in four modes: analyze session corrections into team/project preferences (corrections), extract a single hard-won gotcha into a learned skill (extract), turn a repeatable workflow into a full skill draft (workflow), or passively watch a task as it happens for skill-worthy patterns…
Skill Claude CodeCodex
Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, update or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.