Agent skills you can check before you install
Public SKILL.md skills from GitHub repositories with 50+ stars and from the vendors themselves, scanned and scored. Install any of them with one command in Claude Code, Cursor, Codex, GitHub Copilot or Gemini CLI.
46,100 skills from 2,198 repositories · updated 2 hours ago
9,541 skills
- verification-before-completionSummerEngine/summerRun verification and read its output before claiming work done, fixed or passing: evidence first.OtherNo warnings★ 89↓ —Also in 1 other repo
- decision-reversibilityitseffi/agentic-osClassify decisions as Hat/Haircut/Tattoo by reversibility. Use to calibrate how much analysis a decision deserves.OtherNo warnings★ 117↓ —
- preisangabenThomasMoreAI/legal-skills-openFür Preisangaben (PAngV 2022): ordnet Norm, Beweislast und Gegenargument; Ergebnis: Prüfprodukt mit Risiko und nächstem Schritt.OtherNo warnings★ 96↓ —
- agent-orchestrationmrpulor-gh/nuphusNuphus — 本地优先的 AI Agent：真实桌面执行力 + 手机第二块屏幕。Local-first AI agent with real desktop execution and dual-device real-time sync.OtherNo warnings★ 89↓ —
- quick-win-sessionrshankras/claude-code-apple-skillsGenerates guided first-action flows that help users achieve a meaningful result within 60 seconds to boost retention. Use when user wants quick win onboarding, time-to-value optimization, or first success moments.OtherNo warnings★ 840↓ —
- keel-audittonone-ai/tononeOperational efficiency audit — identify waste, redundancy, and friction across processes, tools, and team workflows. Use when asked to "audit our operations for waste", "where are we inefficient", "what tools are we paying for but not using", or "reduce operational overhead".OtherNo warnings★ 76↓ —
- explain-flagharness/harness-skillsExplain a single FME feature flag in plain language: purpose, default rule, and per-environment targeting state, flagging inconsistencies across environments inline. A quick-lookup/explain capability, not a search across many flags - disambiguates when the name/key given is ambiguous. Doesn't create, kill, or modify anything - see update-flag-targeting for targeting changes, manage-flag-lifecycle for metadata/archive, discover-feature-flags for multi-flag questions, manage-experiments for experiments. Trigger phrases: what does this flag do, explain this flag, how is X configured, is X on in prod, what's targeted for X.OtherNo warnings★ 126↓ —
- tools-internalmrpulor-gh/nuphusNuphus — 本地优先的 AI Agent：真实桌面执行力 + 手机第二块屏幕。Local-first AI agent with real desktop execution and dual-device real-time sync.OtherNo warnings★ 89↓ —
- skill-verwalterThomasMoreAI/legal-skills-openFür Skill-Manager: ordnet Norm, Beweislast und Gegenargument; Ergebnis: Prüfprodukt mit Risiko und nächstem Schritt.OtherNo warnings★ 96↓ —
- verfahrenThomasMoreAI/legal-skills-openFür Umweltrechtliche Verwaltungs- und Gerichtsverfahren: ordnet Norm, Beweislast und Gegenargument; Ergebnis: Prüfprodukt mit Risiko und nächstem Schritt.OtherNo warnings★ 96↓ —
- instrument-metricharness/harness-skillsWire up a track() call for a metric's event, placed at the real outcome, and verify it arrives. Doesn't create the metric (create-metric) or choose it (choose-metric). Trigger phrases: instrument this event, add a track call, wire up tracking, verify the event arrives.OtherNo warnings★ 126↓ —
- agentsclaude-world/director-mode-liteList all available agents (core, expert, self-evolving). Use when the user asks what agents are available or runs /agents.OtherNo warnings★ 80↓ —
- influence-strategiesitseffi/agentic-osGenerate influence strategies using Cialdini's seven principles. Use when you need to influence toward an outcome.OtherNo warnings★ 117↓ —
- duldung-60aThomasMoreAI/legal-skills-openFür Duldung Paragraf 60a: Fachmodul im Migrationsrecht: ordnet Norm, Beweislast und Gegenargument; Ergebnis: Prüfprodukt mit Risiko und nächstem Schritt.OtherNo warnings★ 96↓ —
- release-postermrpulor-gh/nuphusNuphus — 本地优先的 AI Agent：真实桌面执行力 + 手机第二块屏幕。Local-first AI agent with real desktop execution and dual-device real-time sync.OtherNo warnings★ 89↓ —
- zeugenbeistandThomasMoreAI/legal-skills-openFür Zeugenbeistand: ordnet Norm, Beweislast und Gegenargument; Ergebnis: Schnittstellenkarte mit Zuständigkeits- und Nachweisfragen.OtherNo warnings★ 96↓ —
- mcp-toolsmrpulor-gh/nuphusNuphus — 本地优先的 AI Agent：真实桌面执行力 + 手机第二块屏幕。Local-first AI agent with real desktop execution and dual-device real-time sync.OtherNo warnings★ 89↓ —
- shapely-computeparcadei/Continuous-Claude-v3Computational geometry with Shapely - create geometries, boolean operations, measurements, predicatesOtherNo warnings★ 3.9K↓ —Also in 1 other repo
- apple-ecosystem-expertroedyrustam/vibes-plugApple ecosystem development: iOS support, Safari WebKit, and HIG / Pengembangan ekosistem Apple: kompatibilitas iOS, Safari, dan HIG.OtherNo warnings★ 115↓ —
- due-diligenceThomasMoreAI/legal-skills-openFür Kommerzielle Vertrags-DD: ordnet Norm, Beweislast und Gegenargument; Ergebnis: Prüfprodukt mit Risiko und nächstem Schritt. Fachgebiet: Großkanzlei Corporate/M&A.OtherNo warnings★ 96↓ —
- update-flag-targetingharness/harness-skillsChange how a Harness FME feature flag serves traffic in environments. Covers percentage rollout, individual targets, targeting rules (attribute/segment/flag-dependency), defaultTreatment, trafficAllocation, treatments, kill/restore, and copying or initializing a definition in an environment. Use when asked to ramp a flag, add rules/users, change allocation, kill/restore, or promote config. Not for creating flags (create-feature-flag), metadata like tags/owners/archive (manage-flag-lifecycle), segment keys (manage-segments), pipeline rollouts (fme-pipeline), or flag analysis (explain-flag). Trigger phrases: update targeting, ramp flag, percentage rollout, add rule, add users, kill flag, restore, copy config.OtherNo warnings★ 126↓ —
- widget-coherenceserac-labs/seracBuild Service Portal widgets (sp_widget) with synchronized server/client/HTML — data.* initialization, ng-click + controller method matching, c.server.get action handlers, and Angular directive usage.OtherNo warnings★ 77↓ —
- create-templateharness/harness-skillsGenerate Harness v0 Template YAML for reusable pipeline components (Step, Stage, Pipeline, StepGroup) and create them via MCP. Use when asked to create a template, build a step template, stage template, pipeline template, or reusable step. Do NOT use for checking template usage or references (use template-usage instead). Trigger phrases: create template, step template, stage template, pipeline template, reusable step, template YAML.OtherNo warnings★ 126↓ —
- migrate-pipelineharness/harness-skillsMigrate Harness pipelines from v0 to v1 simplified format via MCP tools. Converts expressions from angle-bracket to ${{ }} syntax, lowercases type names, simplifies structure, and preserves variables, failure strategies, and conditional logic. Use when asked to migrate a pipeline, convert v0 to v1, upgrade pipeline format, modernize pipeline syntax, or compare v0 vs v1. Trigger phrases: migrate pipeline, convert to v1, upgrade pipeline, v0 to v1, modernize pipeline, simplified format.OtherNo warnings★ 126↓ —
steady-flame
tmcfarlane/team-avatar
OtherUpdated 3 days ago
From the author
Zuko's steady flame for vetting a performance number. A flame that flickers is not a measurement. Use for benchmark-checklist, /benchmark-checklist, before you report or act on a speedup, a regression, a latency, a throughput, a hillclimb harness, or a library or config choice you measured. Checks the limiter, tuning, physical limits, errors, repeatability, relevance, and whether the work happened at all.
Install
npx skills add tmcfarlane/team-avatar --skill steady-flame -a claude-code
Installs to .claude/skills/steady-flame/ · Each command fetches the skill straight from GitHub with the open skills CLI.
Download .zip
Byte-identical to the scanned commit, with the repository’s licence and a SOURCE.txt.
What it can touch
Instructions onlyNo network use seenLicence: MIT
Safety checks on SKILL.md
No warnings.
Why 83
Security100
Provenance65
Maintenance100
Licence100
Adoption35
Files (1)
- SKILL.md