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.
48,168 skills from 2,325 repositories · updated 7 hours ago
48,168 skills
- constraint-driven-developmentaddyosmani/agent-skillsEstablishes a project's quality bar as a written contract and stops agents quietly lowering it. Interviews the user on which dimensions matter, supplies sane default thresholds when they have no number in mind, records everything in CONSTRAINTS.md, and watches the diff for a weakened bar — new @ts-ignore or eslint-disable suppressions, skipped or deleted tests, assertions stripped out, unimplemented stubs, thresholds edited down. Use when no quality bar is written down, when the user says "set up constraints" or "define our standards", when the user wants dimensions they care about — accessibility, web performance, coverage — set up as enforced constraints, when an agent keeps silencing checks or skipping tests to get to green, when you need a coverage or performance threshold and don't know what number to pick, or when an agent writes more code than anyone will read.TestingNo warnings★ 112.1K↓ —
- hyperliquid-readerhimself65/finance-skillsRead Hyperliquid (app.hyperliquid.xyz) perp and spot market data through opencli and Hyperliquid's public info API: mark, oracle, and mid prices, 24h change, funding rates (hourly and annualized APR), open interest, volume, the L2 order book, OHLCV candles, funding history, and cross-venue funding comparisons (Hyperliquid vs Binance vs Bybit) for funding arbitrage. Use this skill whenever the user mentions Hyperliquid (HL, HL DEX, hyperliquid.xyz) or asks about its perps, spot pairs such as PURR, funding, open interest, order book, or candles. Read-only market data: no account, order, or transfer operations.EngineeringNo warnings★ 3.5K↓ —
- earnings-recaphimself65/finance-skillsAnalyze a company's most recent (or a specified past) earnings report from Yahoo Finance data (yfinance): actual vs estimated EPS, surprise size, revenue and margin trends, and the stock's price reaction. Use this skill whenever the user asks how earnings went — beat or miss, earnings surprise, quarterly results, the post-earnings move, or an earnings call recap — including casual references to a past report such as "AMZN reported last night" or "how did they do". For an upcoming report, use earnings-preview.FinanceNo warnings★ 3.5K↓ —
- closedavepoon/buildwithclaudeCommit the session to durable vault memory and prepare a clean resume pointEngineeringNo warnings★ 3.7K↓ —
- fix-cimichael-denyer/pstack-claudeFind failing PR checks, inspect logs or external check links, and apply focused fixesOtherNo warnings★ 3.7K↓ —
- work-with-prcode-yeongyu/oh-my-openagentFull PR lifecycle in a fresh task-owned git worktree: implement via the ulw-loop skill with mandatory evidence-bound manual QA → reviewer-readable English PR → verification loop (CI + Cubic, where Cubic is skipped only when its quota is exhausted) → merge by default → worktree cleanup. Decomposes one task into the smallest atomic, independently-mergeable PRs and builds the independent ones concurrently via one worktree per PR driven by parallel subagents or a team. Unbounded loop: any failing gate sends you back to fix-and-re-QA inside that PR's worktree. Use whenever implementation work needs to land as a PR. Triggers: 'create a PR', 'implement and PR', 'work on this and make a PR', 'implement issue', 'land this as a PR', 'split into atomic PRs', 'parallel PRs', 'work-with-pr', 'PR workflow', 'implement end to end', even when user just says 'implement X' if the context implies PR delivery.TestingNo warnings★ 70.5K↓ —Also in 1 other repo
- social-selling-planneraaron-he-zhu/aaron-marketing-skillsUse when the user asks to "set up my founder social-selling routine", "build a daily engagement block for target accounts", or "turn funding / hiring signals into selling plays"; produces the founder/seller daily operating block — a time-boxed engagement-block spec (substantive value-add comments on target-account posts, never a pitch), warm-touch-before-ask cadence rules, trigger-response plays consuming the social-pulse-monitor B2B trigger watchlist (funding / hiring / launch signals), and a quarterly diagnostic that reads LinkedIn SSI as an input diagnostic only (Estimated, vendor-defined — never a target metric). All volume norms are Estimated/community-derived; all execution is human — zero mass-DM, zero connection-request automation, zero engagement automation (LinkedIn UA §8.2 red line); 1:1 pitch and DM mechanics route to outreach-manager. Not for cold email sequences — use cold-outbound-sequencer. 社交销售/创始人日常互动块/热身触达节奏/触发信号剧本PlanningNo warnings★ 3K↓ —
- howmichael-denyer/pstack-claudeUse for "how does X work", code walkthroughs before changing something, and placement / ownership / layering questions ("where should this live", "which package owns this", "is this the right layer"). Explains subsystem architecture, runtime flow, onboarding mental models. Use why for motivation.PlanningNo warnings★ 3.7K↓ —
- tddbacknotprop/pstackUse only when the user explicitly asks for TDD, a failing test, or a regression test, OR when the bug has an obvious cheap local test target. Skip when the test path is unclear, expensive, integration-heavy, or not requested.TestingNo warnings★ 3.3K↓ —Also in 3 other repos
- gsd:graphifydavepoon/buildwithclaudeBuild, query, and inspect the project knowledge graph in .planning/graphs/PlanningNo warnings★ 3.7K↓ —
- voice-dossier-builderaaron-he-zhu/aaron-marketing-skillsUse when the user asks to "codify how our founder sounds", "build a founder voice dossier", or "set our content pillars"; runs an 80%-extraction interview over the user's OWN posts, emails, and decks (never competitor scraping, never an invented persona) and produces the versioned voice record — a per-platform register map (incl. 小红书/微信公众号), banned phrases, per-context disclosure lines (the ECHO C2 upstream), a few-shot bank built exclusively from own posts, and 3-5 content pillars with Estimated %-allocations — submitted via memory/events/channels.ndjson for channel-registry to store as voice-dossier.md, the record every Craft-phase skill reads first. Not for the brand-level voice canon — use brand-language-codifier; not for audience/persona research — use audience-mapper. 声音档案/品牌语气/创始人语气/内容支柱/披露声明BrowserNo warnings★ 3K↓ —
- principle-test-behavior-not-implementationbacknotprop/pstackApply when you write, change, or keep a test. Assert what the code's users can observe, so that the test fails when that behavior breaks and keeps passing when only the internals change.TestingNo warnings★ 3.3K↓ —Also in 1 other repo
- nikadavepoon/buildwithclaudeRuns repeatable AI work as checked, budgeted workflow files. The agent captures a repeated task as a .nika.yaml DAG, audits cost/permits/schema before a single token is spent, and runs it with tamper-evident traces.ProductivityNo warnings★ 3.7K↓ —
- adobe-data-handlingjeremylongshore/tons-of-skills-marketplaceAnalyze and enforce classification, minimization, signed-URL custody, retention, deletion, content provenance, and privacy-request boundaries for Adobe workflows. Use when the task requires adobe data custody and privacy review. Trigger with "Adobe data handling", "PDF privacy", or "Firefly content policy".DocumentsNo warnings★ 2.9K↓ —
- skyverndavepoon/buildwithclaudeAI-powered browser automation — navigate sites, fill forms, extract structured data, log in with stored credentials, and build reusable multi-step workflows using natural language. Install: pip install skyvern && skyvern setupBrowserNo warnings★ 3.7K↓ —
- doca-aes-gcmNVIDIA/skillsUse this skill when the user is doing hands-on DOCA AES-GCM work on a BlueField DPU or ConnectX NIC — configuring `doca_aes_gcm_task_encrypt` / `_task_decrypt`, querying `doca_aes_gcm_cap_*` for per-key-type (only `DOCA_AES_GCM_KEY_128` / `_256` — AES-192 not supported) and per-task support, sizing plaintext against the max-buf cap, setting source / destination mmap permissions, validating with a NIST GCMVS or RFC 5288 vector, or debugging DOCA_ERROR_* including the security-critical tag-verification-failed outcome on decrypt. Trigger even when the user does not explicitly mention "DOCA AES-GCM" or "AEAD" — typical implicit phrasings: "decrypt completion IO_FAILED", "auth tag isn't verifying", "NOT_PERMITTED on my encrypt buffer", "is AES-192-GCM on this BlueField" (no), or "encrypted record came back tampered". Refuse and route elsewhere for non-GCM AES modes (CBC / CTR / XTS — CPU OpenSSL), key management (KMS / HSM / rotation), SHA (doca-sha), or general AEAD background.SecurityNo warnings★ 3.8K↓ —
- get-pr-commentsmichael-denyer/pstack-claudeFetch and summarize review comments from the active pull requestEngineeringNo warnings★ 3.7K↓ —
- running-load-testsjeremylongshore/tons-of-skills-marketplaceCreate and execute load tests for performance validation using k6, JMeter, and Artillery. Use when validating application performance under load conditions or identifying bottlenecks. Trigger with phrases like "run load test", "create stress test", or "validate performance under load".TestingNo warnings★ 2.9K↓ —
- canva-cost-tuningjeremylongshore/tons-of-skills-marketplaceReduce unnecessary Canva Connect operations using measured request, retry, cache, and premium-feature evidence. Use when reviewing usage, trial-quota pressure, deduplication, or budget guardrails. Trigger with: "reduce Canva usage", "Canva quota planning", "optimize Canva calls".PlanningNo warnings★ 2.9K↓ —
- conversion-signal-qaaaron-he-zhu/aaron-marketing-skillsUse when the user asks to "QA my conversion tracking before launch", "check my UTMs / pixel / event firing", "set up a tracking pre-flight", or "set the dedup rule so Meta and Google stop double-counting"; builds and fixes the measurement plumbing — conversion-event firing, UTM hygiene, cross-platform dedup rules, attribution-window alignment, and offline/iOS-ATT modeled-gap flags — as a pre-flight checklist plus a UTM/event-spec builder. Not for scoring R1/R2 — that is a scored veto in ad-account-auditor; not for account structure — use campaign-architect. 付费广告转化追踪QA/UTM规范/跨平台去重TestingNo warnings★ 3K↓ —
- vercel-hello-worldjeremylongshore/tons-of-skills-marketplaceCreate a minimal working Vercel deployment with a serverless API route. Use when starting a new Vercel project, testing your setup, or learning basic Vercel deployment and API route patterns. Trigger with phrases like "vercel hello world", "vercel example", "vercel quick start", "simple vercel project", "first vercel deploy".TestingNo warnings★ 2.9K↓ —
- constant-time-testingtrailofbits/skillsMeasures timing side channels in cryptographic implementations by running them, using dudect for statistical analysis and Timecop over Valgrind for dynamic tracing. Covers the formal, symbolic, dynamic, and statistical tool categories and how to read a result. Use when testing whether a running implementation is constant-time, measuring timing variance on a compiled binary, or investigating a suspected timing attack. Not for statically inspecting compiler output — the constant-time-analysis plugin covers that.TestingNo warnings★ 7.9K↓ —
- launch-day-conductoraaron-he-zhu/aaron-marketing-skillsUse when the user asks to "run my launch day", "build a launch day runbook / war room", or "decide CONTINUE or ROLLBACK after the push"; produces a pre-conditions gate check (launch-readiness-auditor SHIP verdict + the authoritative date in launch-registry — missing either stops the skill), a dated hour-blocked runbook with owners (morning irreversible pushes, daytime monitoring loop, evening consolidation), a forced observation-window verdict after every irreversible action against pre-declared kill criteria, a P0-P3 incident ladder with rollback playbooks, and T-0 status lines for the registry proposal protocol. Not for channel submission content and platform rules — use community-launch-runner; not for media replies — use press-media-relations. 发布日runbook/作战室/观察窗/回滚裁决/发布日指挥OtherNo warnings★ 3K↓ —
- navan-data-syncjeremylongshore/tons-of-skills-marketplaceSynchronize Navan booking, expense, identity, and file-delivered data according to each documented source contract. Use when building durable ETL or correcting sync drift. Trigger with "Navan data sync", "Navan ETL", or "reconcile Navan feeds".DataNo warnings★ 2.9K↓ —
tao-generate-anomalies
NVIDIA/skills
OtherUpdated yesterday
From the author
Full 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.
Install
npx skills add NVIDIA/skills --skill tao-generate-anomalies -a claude-code
Installs to .claude/skills/tao-generate-anomalies/ · 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: APACHE-2.0
Safety checks on SKILL.md
No warnings.
Why 90
Security100
Provenance83
Maintenance100
Licence100
Adoption57
Files (22)
- BENCHMARK.md
- SKILL.md
- assets/ag_config.yaml
- assets/defect_spec_template.jsonl
- config/skillspector-baseline.yaml
- evals/evals.json
- references/datasets.md
- references/docker.md
- references/error-handling.md
- references/eval.md
- references/finetune-commands.md
- references/finetune.md
- +10 more