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.
49,409 skills from 2,375 repositories · updated 56 minutes ago
49,407 skills
- rag-evalNVIDIA/skillsFilesystem RAG benchmarks: corpus/, train.json, evaluate_rag.py (RAGAS quality). Not for prod monitoring, latency/throughput benchmarking (use rag-perf), or evals outside this repo layout.OtherNo warnings★ 3.8K↓ —
- kiro-debuggotalab/cc-sddInvestigate implementation failures using root-cause-first debugging. Use when an implementer is blocked, verification fails, or repeated remediation does not converge.EngineeringNo warnings★ 3.8K↓ —Also in 7 other repos
- kiro-spec-requirementsgotalab/cc-sddGenerate EARS-format requirements based on project description and steering context. Use when generating requirements from project description.PlanningNo warnings★ 3.8K↓ —
- pre-publish-reviewcode-yeongyu/oh-my-openagentNuclear-grade 12-agent pre-publish release gate. Runs /get-unpublished-changes to detect all changes since last npm release, spawns up to 10 ultrabrain agents for deep per-change analysis, invokes /review-work (orchestrator manual QA plus one gate reviewer) for holistic review, and 1 oracle for overall release synthesis. Runs ONLY when the user explicitly asks for a pre-publish review — a plain publish/release request MUST NOT trigger this; /publish ships directly. Triggers: 'pre-publish review', 'review before publish', 'release review', 'pre-release review', 'ready to publish?', 'can I publish?', 'pre-publish', 'safe to publish', 'publishing review', 'pre-publish check'.EngineeringNo warnings★ 70.7K↓ —Also in 1 other repo
- tao-generate-anomaliesNVIDIA/skillsFull PAIDF AnomalyGen pipeline — fine-tune on a new anomaly dataset, generate synthetic anomaly images (SDG), evaluate quality (nn_score), and search per-sample (guidance, crop_ratio) parameters. Three modes: full (Phase 0→7: finetune then generate), finetune_only (Phase 0→1: train only), inference_only (Phase 0, 2→7: generate from an existing checkpoint). Use when the user asks to "fine-tune AnomalyGen", "generate anomaly images", "run PAIDF SDG", "evaluate SDG output quality", "run per-sample search", or run any part of the AnomalyGen pipeline, even if they only mention one phase.OtherNo warnings★ 3.8K↓ —
- paid-ads-launchnowork-studio/notfair-pluginPlan and prepare a new paid-media campaign or cross-channel test before it can spend. Use when asked to create, launch, start, set up, or allocate budget to an advertising campaign on any supported platform.MarketingNo warnings★ 4.1K↓ —
- ops-marketingdavepoon/buildwithclaudeMarketing command center. Email campaigns (Klaviyo), paid ads (Meta/Google), analytics (GA4), SEO, and social media metrics. One dashboard for all marketing channels.MarketingNo warnings★ 3.7K↓ —
- kiro-validate-designgotalab/cc-sddInteractive technical design quality review and validation. Use when reviewing design before implementation.DesignNo warnings★ 3.8K↓ —
- kiro-steering-customgotalab/cc-sddCreate custom steering documents for specialized project contexts. Use when creating domain-specific steering files.PlanningNo warnings★ 3.8K↓ —
- debriefdavepoon/buildwithclaudeAlias for `/origin:handoff` — symmetric brief/debrief naming. Same behavior: end-of-session ritual that writes session log + project status + granular MCP captures. Invoked as `/debrief`. Use when the user prefers the brief/debrief pair over brief/handoff.WritingNo warnings★ 3.7K↓ —
- doca-perftestNewNVIDIA/skillsUse this skill when the user is running DOCA Perftest (doca_perftest), the RDMA benchmark that measures bandwidth, message rate and latency for RDMA verbs, point-to-point or across a cluster, with an optional DOCA Verbs backend (-r dv) or GPUNetIO engine (--engine gpunetio). Trigger even when the user does not say "DOCA Perftest" — implicit phrasings include "how do I benchmark RDMA write bandwidth with DOCA", "run ib_write_bw but across 8 nodes", "perftest says the device is busy", "my latency numbers differ between client and server", "launch the server over SSH", or "run an all-to-all RDMA test from a JSON scenario". Refuse and route elsewhere for DOCA install (doca-setup), DOCA Bench (doca-bench), and application-level throughput.TestingNo warnings★ 3.8K↓ —
- eval-integritycode-yeongyu/oh-my-openagentAssess integrity of an eval, benchmark, metric, A/B test, holdout, LLM-judge, or user-study claim when the question is whether measurement, comparison, or evidence is independent and trustworthy.TestingNo warnings★ 70.7K↓ —
- pi-history-ingestAr9av/obsidian-wikiIngest Pi coding-agent session history into Obsidian as distilled knowledge. Use for importing or mining ~/.pi/agent/sessions or Pi conversation logs; use wiki-agent for targeted topic-only recall.EngineeringNo warnings★ 3.7K↓ —
- kiro-spec-batchgotalab/cc-sddCreate complete specs (requirements, design, tasks) for all features in roadmap.md using parallel sub-agent dispatch by dependency wave.PlanningNo warnings★ 3.8K↓ —
- node-zombie-guardianforyourhealth111-pixel/Vibe-SkillsUse when diagnosing stale or orphaned Node.js processes launched by VCO, auditing ownership/liveness, or safely simulating cleanup without touching external Node workloads.OtherNo warnings★ 3.8K↓ —
- prisma-postgresprisma/skillsDeprecated compatibility name for prisma-postgres-setup. Use only when a user or existing integration explicitly invokes prisma-postgres; otherwise select prisma-postgres-setup.OfficialDataNo warnings★ 79↓ —
- doca-commonNVIDIA/skillsUse this skill whenever the user is doing hands-on DOCA programming on a BlueField DPU or ConnectX NIC and needs the foundation primitives every per-library context rests on — walking the doca_ctx lifecycle, discovering doca_dev / doca_devinfo and gating on doca_*_cap_* before trusting a feature, wiring doca_mmap / doca_buf_inventory / doca_buf for zero-copy I/O across libraries, driving doca_pe for completions, or DOCA Log's two-tier (--sdk-log-level vs app-side) model. Trigger even when the user does not say "DOCA Common" — typical implicit phrasings include "my tasks submit but nothing completes", "DOCA_ERROR_BAD_STATE from doca_ctx_start", "--sdk-log-level does nothing for my DOCA_LOG_DBG lines", "share a buf between doca_dma and doca_rdma", or "crashes far from the offending line". Refuse and route elsewhere for per-library questions in isolation (load doca-flow / doca-rdma / doca-eth alongside), installing DOCA (doca-setup), or doc lookup (doca-public-knowledge-map).ProductivityNo warnings★ 3.8K↓ —
- social-pulse-monitoraaron-he-zhu/aaron-marketing-skillsUse when the user asks to "monitor brand mentions", "set up social listening", "did anything spike about us this week", or "watch these accounts for buying triggers"; runs always-on keyless listening — a versioned listening-query architecture (brand variants incl. misspellings and 中文 names, exclusion terms, per-source syntax for HN Algolia / Bluesky / GDELT / Tavily), a mention sweep with six-class triage (crisis / bug / lead / praise / question / spam) each routed with an SLA, a 7-day rolling baseline with spike flags (the crisis trigger input), a B2B trigger watchlist (funding / hiring / launch signals), and an explicit coverage disclosure where X/IG/TikTok/LinkedIn/小红书 numbers are proxy-labeled, never Measured. Not for launch-window telemetry (T-0→T+30 rank and review polling) — use launch-monitor. 舆情监听/品牌提及监测/基线突刺/触发信号MarketingNo warnings★ 3.1K↓ —
- principle-experience-firstmichael-denyer/pstack-claudeApply when product, UX, or feature-scope tradeoffs come up. Choose user delight over implementation convenience; ship fewer polished features over more rough ones.DesignNo warnings★ 3.9K↓ —
- hubspot-automationdavepoon/buildwithclaudeAutomate HubSpot CRM operations (contacts, companies, deals, tickets, properties) via Rube MCP using Composio integration.OtherNo warnings★ 3.7K↓ —
- kiro-debuggotalab/cc-sddInvestigate implementation failures using root-cause-first debugging. Use when an implementer is blocked, verification fails, or repeated remediation does not converge.EngineeringNo warnings★ 3.8K↓ —
- jetson-video-pipelineNVIDIA/skillsUse when planning, executing, and independently validating Jetson Video Codec SDK or PyNvVideoCodec encode/decode, transcode, segmentation, container decode, AV1, or concise acceptance workflows.MediaNo warnings★ 3.8K↓ —
- offsite-signal-analyzeraaron-he-zhu/aaron-marketing-skillsUse when the user asks to "analyze backlinks", "analyze my off-site signals", or "track AI traffic / ChatGPT / Perplexity referrals"; profiles referring domains, anchor-text mix, toxic links, and disavow candidates (backlinks mode) or isolates AI-assistant referral sessions in GA4/GSC/logs and reports their trend, landing pages, and AI-vs-organic conversion (ai-referrals mode). Not for internal links — use site-structure-optimizer; not for keyword positions — use rank-tracker; not for multi-metric stakeholder reports — use performance-monitor. 外链分析/反向链接/AI流量/AI引荐/ChatGPT流量/AI转化MarketingNo warnings★ 3.1K↓ —
- gsd:discuss-phasedavepoon/buildwithclaudeGather phase context through adaptive questioning before planning. Use --all to skip area selection and discuss all gray areas interactively. Use --auto to skip interactive questions (Claude picks recommended defaults). Use --chain for interactive discuss followed by automatic plan+execute. Use --power for bulk question generation into a file-based UI (answer at your own pace).DesignNo warnings★ 3.7K↓ —
principle-build-the-lever
backnotprop/pstack
WritingUpdated yesterday
From the author
Apply to any non-trivial work, not just bulk work: edits, migrations, analyses, checks. Build the tool that does it or proves it (codemod, script, generator, or a skill your subagents follow) instead of working by hand. The tool is the artifact a reviewer can rerun.
Install
npx skills add backnotprop/pstack --skill principle-build-the-lever -a claude-code
Installs to .claude/skills/principle-build-the-lever/ · 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 90
Security100
Provenance82
Maintenance100
Licence100
Adoption56
Files (1)
- SKILL.md