Claude Skills
The open format is called Agent Skills and works in Claude Code, Codex, Cursor and other agents — most people know it as Claude Skills.
Every Agent Skill we could find on GitHub, deduplicated by content. 86 961 files from 2 125 authors, of which 68 678 are unique — the rest is the same skill repackaged into someone else's repository. For each one: what it weighs in tokens, whether it ships runnable scripts, and which MCP servers it needs.
Reconcile App Store Connect subscriptions and in-app purchases with RevenueCat products, entitlements, offerings, and packages using asc and RevenueCat MCP. Use when setting up or syncing subscription catalogs across ASC and RevenueCat.
Resize and validate App Store screenshots with current asc screenshot-size data and macOS sips. Use when preparing or fixing screenshots for App Store Connect submission.
Orchestrate iOS screenshot automation with xcodebuild/simctl for build-run, AXe for UI actions, JSON settings and plan files, Koubou-based framing (`asc screenshots frame`), and screenshot upload (`asc screenshots upload`). Use when users ask for automated screenshot capture, AXe-driven simulator flows, frame composition, or screenshot-to-upload pipelines.
Set up bundle IDs, capabilities, signing certificates, provisioning profiles, and encrypted signing sync with the asc cli. Use when onboarding a new app, rotating signing assets, or sharing them across a team.
Diagnose App Store submission blockers and operate review health with asc, including readiness validation, repair routing, status monitoring, cancellation, and retry decisions. Use when validation fails, a version is not in a valid state, review status is unclear or stuck, or a failed submission must be repaired and retried. For staging, upload, publication, and submission execution, use asc-release-flow.
Bulk-localize subscription, subscription-group, and in-app purchase display names across App Store locales using asc, including API 4.4.1 version-scoped v2 resources. Use when filling or updating subscription/IAP names and descriptions without App Store Connect UI work.
Orchestrate TestFlight distribution, groups, testers, and What to Test notes using asc. Use when rolling out betas.
Submit or update a Wall of Apps entry in the App-Store-Connect-CLI repository using `asc apps wall submit`. Use when the user says "submit to wall of apps", "add my app to the wall", or "wall-of-apps".
Generate engaging, localized App Store release notes (What's New) from git log, bullet points, or free text using canonical metadata under `./metadata`. Optionally pairs with promotional text updates.
Define, validate, run, resume, and audit repo-local multi-step automations with current `asc workflow` and `.asc/workflow.json`, including step outputs and safe release/TestFlight workflows.
Build, archive, generate export options, export, upload, and manage Xcode version/build numbers with the current asc xcode helpers before App Store Connect upload or submission. Use when creating an IPA or PKG for upload.
将约 5 秒口播文稿、观点句或抽象概念做成高级 editorial halftone paper-collage / 半调纸拼贴 B-roll。用户说“collage b-roll”“纸拼贴 b-roll”“半调拼贴”“拼贴风格配画面”“用这段文稿做拼贴动画”“gbro-collage-broll”，或希望把一句文稿转成拼贴视觉隐喻时，必须使用此 skill。强制采用三阶段审批：先只提视觉隐喻，用户确认后才生成彩色拼贴静帧，静帧再次确认后才默认调用 Gemini Omni Flash 生成首尾帧组装动画。默认视频模型固定为 gemini-omni-flash-preview，不再默认使用 Veo；只有用户明确指定其他模型时才切换。
Use this skill when a dependabot dependency update PR fails.
Give an AI agent persistent, cross-session long-term memory through MindMemOS. Covers installing and authenticating the mindmemos CLI, the full command interface (add / search / get / update / delete / feedback / dreaming) with parameters and examples, guidance on which capability to use when, plus a Python SDK example. To wire memory into a specific agent host (OpenClaw, DeepSeek Harness, Codex, Claude, etc.), see references/.
Use this skill whenever the user wants to do anything with Excel spreadsheet files (.xlsx, .xls, .csv). This includes reading data, writing formulas, manipulating cells, formatting, filtering, creating charts, pivot tables, and any spreadsheet automation tasks.
Fix broken dependency update PRs and aggregate the ones that work into one PR.
>- Create a visual Jira sprint dashboard from Jira project, space, sprint, board, filter, JQL, work item keys, or Jira URL data. Use when the user asks for a Jira sprint dashboard, standup dashboard, sprint review, delivery review, engineering manager dashboard, WIP review, planning view, closeout view, or a visual snapshot of Jira work that is more useful than a flat report. Use the richest dashboard format supported by the current agent, such as Cursor Canvas, an interactive artifact, HTML, or Markdown.
Maintain Wassette's Keep a Changelog-style CHANGELOG.md by reviewing a focused change, deciding whether it affects users, and updating the [Unreleased] section with concise, categorized, non-duplicate entries.
When a symptom has several plausible causes, rank falsifiable hypotheses and run the cheapest discriminating observation first; prefer least-assumptive survivors only after evidence fit.
When a symptom has several plausible causes, rank falsifiable hypotheses and run the cheapest discriminating observation first; prefer least-assumptive survivors only after evidence fit.
When a symptom has several plausible causes, rank falsifiable hypotheses and run the cheapest discriminating observation first.
When a symptom has several plausible causes, rank falsifiable hypotheses and run the cheapest discriminating observation first; prefer least-assumptive survivors only after evidence fit.
When a symptom has several plausible causes, rank falsifiable hypotheses and run the cheapest discriminating observation first; prefer least-assumptive survivors only after evidence fit.
When a symptom has several plausible causes, rank falsifiable hypotheses and run the cheapest discriminating observation first; prefer least-assumptive survivors only after evidence fit.
When a symptom has several plausible causes, rank falsifiable hypotheses and run the cheapest discriminating observation first; prefer least-assumptive survivors only after evidence fit.
When a symptom has several plausible causes, rank falsifiable hypotheses and run the cheapest discriminating observation first; prefer least-assumptive survivors only after evidence fit.
Use when search or investigation could run forever. Set an explicit good-enough threshold first, then stop at the first option that clears it.
Use when a specific claim may lack grounding. Check evidence boundary, size wrongness cost, then answer, fetch, or abstain — never confabulate.
When a fault is localized and the proximate cause is known but the systemic root is not, chain evidence-linked whys with a counterfactual stop and a countermeasure.
When the right response mode is unclear, classify the cause-effect domain first; decompose disorder.
When a constraint is treated as fixed, separate physics from convention, keep only independently supported primitives, and rebuild the simplest solution that satisfies real constraints.
Deciding what to build or why adoption fails. Recover the progress users hire a solution for under a circumstance, then rank by outcome and competing workarounds.
Use when a selective defect needs IS/IS-NOT difference analysis or a consequential option choice needs must/want weighting and adverse-consequence comparison.
Under genuine uncertainty with no reliable forecast, inventory means, cap downside at affordable loss, act for commitments, and let goals emerge from controllable moves.
Use when longevity of a non-perishable option matters. Treat survival duration as a remaining-life prior, then check domain drift before favoring the proven.
When a claim, doc, test, metric, or assumption conflicts with observed behavior, stop theorizing from the map and verify the live code or data; let territory overrule.
When one mental model leaves a material blind spot on a multi-domain or high-stakes problem, sequence complementary models with named roles and a conflict rule.
Use under time pressure when the situation is still changing and you must act before certainty — cycle Observe→Orient→Decide→Act on ~70% confidence, then re-observe.
When unsure which thinking skill fits, map domain and problem type, then return NONE or one primary skill by default (at most three complementary).
Before committing scarce time, people, or money, name the best forgone use of those resources and the value delta of the chosen path versus that alternative.
When provisioning, setting a limit, or committing an estimate under uncertainty, size a buffer to residual error and the cost of breach—not to the optimistic edge.
Before committing to a plan or launch, assume it already failed and reason backward through concrete causes — convert failure paths into mitigations, gates, and stop checks.
Use when forecasting, estimating, or sizing risk — anchor on base rates, give ranges, update prior→likelihood→posterior on evidence, and factor unmeasured quantities into order-of-magnitude bounds.
For authorized security review of code, auth, or APIs you control, model the attacker, map the attack surface, and report only findings with a reproducible exploit path and verified mitigation.
Before heavy deliberation, classify the decision as cheap or costly to undo; decide two-way doors fast and stage one-way doors to preserve options.
When a change has effects past the immediate fix—incentives, scale, feedback—trace consequence chains with timing and probability before committing.
When a request is vague, assumption-laden, or "obvious," ask the few load-bearing questions that expose hidden requirements before building or committing.
Before rejecting a proposal or reflexively agreeing, build the strongest faithful opposing case, state agreement conditions, then update or reaffirm.
When behavior is emergent across components—fixes elsewhere break, loops/delays dominate—map boundary, stocks/flows, feedback, archetypes, then rank leverage.
When a real test is too rare, large, or irreversible, run a controlled counterfactual: isolate one variable, fix conditions, trace the mechanistic chain, and bound what the result implies.
When throughput or latency is pipeline-limited, identify the single binding constraint and exploit, subordinate, elevate, then recheck—ignore non-constraints.
Use when the reflex is to add a feature, layer, or process. Prefer removing harmful or nonessential elements first, with an irreversibility guard before deletion.
When two design requirements seem mutually exclusive, name the contradiction, separate conflicting states, then invent a concrete no-compromise resolution.
Use Claude Code as an independent `claude -p` subagent when the user explicitly asks for Claude, wants a second-agent opinion from Claude, or asks to delegate a well-scoped task to Claude. Supports selecting `--model` and thinking/effort level with defaults of `opus` and `high`.
Audit the choices an implementing agent made, not its diff — a pure decision audit that traces the session's history into a choices ledger, changes no code, and never blocks an unsupervised run. Working code still embeds architecture the user never chose; surface it because future work inherits it. Use when the user wants to review the decisions the AI made on their behalf, before merging or committing AI-implemented work, when integrating a delegated subagent's pass, or when a fix "works" but might be a point fix.
Run fast, progressive experiments to map tunable parameters and their effects. Use when optimizing code, prompts, configurations, or other artifacts against a goal or benchmark, or when a long research run needs a planned hypothesis queue, focused trials, and durable findings.
Review changed code for naming, stale references, unnecessary complexity, and comment quality. Use after completing implementation work, before committing, or when the user asks to review or audit code.
Eval and improve a skill against golden cases — run the target skill blind in a fresh, context-free subagent on each example input, grade the artifact against the expected outcome, and let the gaps drive the edits. Use when the user wants to test/eval/improve/harden a skill, says "this skill keeps producing X / keeps missing Y", or hands a skill plus example input→expected-output pairs. Pairs with [write-skills](../write-skills/SKILL.md) (the authoring principles every fix obeys).
Create or revise agent skills. Use when adding a new skill file, renaming a skill, simplifying an existing skill, improving trigger descriptions, or deciding what belongs in a skill versus references, scripts, assets, or ordinary docs.
When a spec under specs/<feature>/ is done shipping (a write-spec build, or any planned task that produced a spec), archive it to specs/done/ and rewrite it from a build-plan into a durable rationale — the why, the principles, the invariants — pointing back to the real code for the how. Use when implementation has landed and the plan no longer matches what shipped, or the user says a feature/spec is finished. Pairs with [write-spec](../write-spec/SKILL.md) (the plan this closes) and [review](../review/SKILL.md) (run before closing).
Claude Skills — questions
Answers built from the skills we actually parsed.