Loading...
Loading...
Found 3,217 Skills
Analyze an unfamiliar codebase and generate a structured onboarding guide with architecture map, key entry points, conventions, and a starter CLAUDE.md. Use when joining a new project or setting up Claude Code for the first time in a repo.
Design MVPs, validated learning experiments, and pivot-or-persevere decisions using Build-Measure-Learn. Use when the user mentions "MVP scope", "validated learning", "pivot or persevere", "vanity metrics", or "test assumptions". Covers innovation accounting and actionable metrics. For 5-day prototype testing, see design-sprint. For customer motivation analysis, see jobs-to-be-done. Trigger with 'lean', 'startup'.
Critique-and-rewrite enforcement loop for voice fidelity. Validates generated content against negative prompt checklists and forces revision until it passes. Use when content has been generated in a target voice, voice output feels off, long-form content risks voice drift, or before final delivery of voice content. Use for "validate voice", "check voice", "voice feels wrong", "voice drift", or "rewrite for voice". Do NOT use for initial voice generation, voice profile creation, or content that has no voice target.
Collaborative coding with enforced micro-steps: announce, show diff, wait for confirmation, apply, verify. User controls pace with commands. Works with any domain agent as the executor. Use when: "pair program", "pair with me", "let's code together", "step by step coding", "walk me through implementing", "code with me"
Merge validated feature to main via PR, tag release, cleanup worktree. Use after /feature-validate passes. Use for "release feature", "merge feature", "ship it", or "/feature-release". Do NOT use without passing validation or for hotfixes that skip the pipeline.
Reorganize the user's X and LinkedIn network with review-first pruning, add/follow recommendations, and channel-specific warm outreach drafted in the user's real voice. Use when the user wants to clean up following lists, grow toward current priorities, or rebalance a social graph around higher-signal relationships.
Deep single-word English mastery skill. Unpacks one English word's semantic image, modern usage, confusion points, and memory hook. Use when the user wants to deeply understand one English word, asks for deeper meaning or usage rather than translation, or wants a memorable vocabulary breakdown. Prefer this over `plain` or `learn` for a single vocabulary item. Do not use it for phrases, sentence translation, or multiple words at once; route multi-word card workflows to `word-flow`.
CEO/Founder-mode plan review. Rethink the problem, find the 10-star product, challenge premises, expand scope when it creates a better product. Four modes: SCOPE EXPANSION (dream big), SELECTIVE EXPANSION (hold scope + cherry-pick), HOLD SCOPE (maximum rigor), SCOPE REDUCTION (strip to essentials). Adapted from gstack/plan-ceo-review (Garry Tan, MIT license).
Get AI-powered match predictions for Premier League and Champions League including scores, next goal, and corners.
Audit experiment integrity before claiming results. Uses cross-model review (GPT-5.4) to check for fake ground truth, score normalization fraud, phantom results, and insufficient scope. Use when user says "审计实验", "check experiment integrity", "audit results", "实验诚实度", or after experiments complete before writing claims.
Run a formal, multi-dimensional code review of a pull request. Reads the PR diff, classifies change types, dispatches parallel reviewers by dimension (correctness, consistency, docs-sync, plus conditional security/edge-cases/UX/performance/structure/maintainability), and synthesizes findings into an actionable punch list. Use when the user asks to review a PR, run /deep-review, mark a PR as ready for review, or requests a formal/thorough code review.
Plan how to slice a non-trivial coding task across parallel subagents. Returns a dispatch plan (file assignments, dependencies, output-format contracts) — the main Agent then executes it with the Agent tool + `isolation: "worktree"`. Invoke only when work justifies multi-agent overhead: (a) greenfield 0→1 across multiple independent modules, (b) change touches ≥3 modules, or (c) ≥5 files each with >50 lines of diff. Small changes write inline.