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.
48,015 skills from 2,309 repositories · updated 31 minutes ago
6,173 skills
- ai-git-cleanuparcasilesgroup/ai-engineeringRemoves finished local git work in one automatic pass, with no question and no name to type: dead worktrees first (`git worktree remove`, then `git worktree prune`), then branches already merged into the local `main`, then the branches a squash-merged pull request left behind (tree identical to `origin/main`, tagged before deletion). Trigger for "tidy up", "tidy up these branches", "tidy branches", "delete old branches", "start fresh", "limpieza de branches", "limpia branches", or by naming ai-git-cleanup. A branch that is not merged into `main` is left alone — unless its tree is byte-identical to `origin/main`, which is what a squash-merged pull request looks like afterward; the pass deletes those with the proof in hand, tags the tip, and never force-deletes anything else. Not for pushing, not for remote branches, not for opening a pull request.EngineeringNo warnings★ 62↓ —
- ai-adversarial-looparcasilesgroup/ai-engineeringCoordinate a multi-round debate between two agents until the code is clean. adversarial-reviewer (the critic) assumes everything is wrong; fixer implements fixes or disputes them with evidence. They exchange findings through a shared thread file, and each keeps its memory across rounds. Use as the orchestrator's review gate, or when the user asks for an adversarial review-and-fix loop on recent changes (e.g. "/ai-adversarial-loop", "have the critic and fixer go at checkpoint 3").EngineeringNo warnings★ 62↓ —
- review-routerarcasilesgroup/ai-engineeringDecide which review lanes a change actually needs, then run exactly those — instead of spawning the whole chain at every diff size. Reads the diff's shape and blast radius, maps the evidence to lanes (code, security, a11y, perf, design, build), estimates the cost, shows the routing decision, and executes it. Use when asked to "review this", "check this before I commit/merge", "is this ready", or "what should I run", and whenever a full review would cost more than the change is worth. Prefer this over full-review as the default entry point; full-review remains correct when the user explicitly asks for everything.EngineeringNo warnings★ 62↓ —
- scenario-godot-2dUpdatedscenario-labs/skillsUse when building a 2D game in Godot 4.7: make a platformer or top-down controller that feels right (coyote time, jump buffer, jump height), TileMapLayer and TileSet in code, terrains and autotiles that pick wrong tiles, a dual grid, a camera that jitters or stutters, pixel art that looks blurry or uneven, stretch and integer scaling, 2D lights and shadows, y-sort order wrong, parallax gaps, or importing AI-generated sprites and tilesets from Scenario.EngineeringNo warnings★ 1.2K↓ —
- slack-mrkdwnccheney/robust-skillsFormat or debug Slack message text, mentions, links, dates, and escaping. Use when choosing Slack mrkdwn versus Markdown, rich_text, or plain_text for a receiving field; not Block Kit layout or Slack history retrieval.EngineeringNo warnings★ 61↓ —
- full-reviewarcasilesgroup/ai-engineeringRun every review skill at once — build-check, then design-check, code-audit, security-audit, a11y-audit and perf-audit fanned out across parallel agents — and merge them into one deduplicated, severity-ranked report. Use when asked for a full/complete review, to review everything before shipping or merging, or to run all the checks.EngineeringNo warnings★ 62↓ —
- slack-block-kitccheney/robust-skillsBuild or debug Slack Block Kit payloads for messages, modals, App Home, streaming responses, and Work Object unfurls. Use for block layout, elements, or interaction schemas; not plain Slack text formatting alone.EngineeringNo warnings★ 61↓ —
- clerk-cliclerk/skillsOperate the Clerk CLI (`clerk` binary) for authentication, user/org/session management, impersonation, local webhook testing, deploy verification, instance config, env keys, feature toggles, and any Clerk Backend, Platform, or Frontend API call. Use when the user mentions Clerk management tasks, "list clerk users", "impersonate a user", "test webhooks locally", "enable orgs", "enable billing", "clerk env pull", "clerk doctor", "clerk deploy", "clerk api", or any ad-hoc Clerk API request. Prefer the CLI over raw HTTP: it handles auth, key resolution, app/instance targeting, and formatting automatically.OfficialEngineeringReview first★ 89↓ —
- nemo-guardrailsKalarisLabs/research-agent-skillsAdds runtime safety rails to LLM applications with NVIDIA NeMo Guardrails, configured through Colang 2.0 flows. Covers jailbreak and prompt-injection detection, self-check input/output validation, retrieval-based fact-checking, hallucination detection, PII filtering via Presidio, toxicity detection via ActiveFence, and LlamaGuard integration. Use when adding programmable safety rules to a production LLM app, blocking jailbreaks or prompt injection, filtering PII from model inputs and outputs, verifying generated claims against retrieved sources, or tuning false positives and latency of guardrail checks. For standalone moderation only, use LlamaGuard or the OpenAI Moderation API instead.EngineeringNo warnings★ 79↓ —
- evaluating-code-modelsKalarisLabs/research-agent-skillsEvaluates code generation models across HumanEval, MBPP, MultiPL-E, and 15+ benchmarks with pass@k metrics. Use when benchmarking code models, comparing coding abilities, testing multi-language support, or measuring code generation quality. Industry standard from BigCode Project used by HuggingFace leaderboards.EngineeringNo warnings★ 79↓ —
- better-github-skillAVGVSTVS96/better-github-skillGitHub work via gh CLI: PRs, review conversations, CI failures, repo state. Use when inspecting a PR, reading a review (bodies, comments, inline threads), debugging CI (finding which run failed, why a check or Actions run is red, reading job logs), or composing non-trivial gh commands. Scripts cover what raw gh can't do (thread resolution state) or reliably fumbles (PR snapshots, CI log drilldown); use raw gh directly for everything else.EngineeringNo warnings★ 52↓ —
- devexpress-office-file-api-presentationDevExpress/agent-skillsBuild .NET applications with the DevExpress Presentation API for creating, reading, modifying, and exporting PowerPoint presentations programmatically without Microsoft Office. Use when working with .pptx files, slides, shapes, text, tables, slide masters and layouts, themes, slide backgrounds, notes, headers/footers, or exporting presentations to PDF. Also use when someone mentions "DevExpress Presentation", "Presentation API", "DevExpress.Docs.Presentation", "create PowerPoint in C#", "pptx automation .NET", "merge presentations", or asks about any PowerPoint/presentation processing with DevExpress. Covers .NET.EngineeringNo warnings★ 67↓ —
- devexpress-office-file-api-barcodeDevExpress/agent-skillsBuild .NET applications with the DevExpress Barcode Generation API for generating barcode images programmatically. Use when generating QR codes, Data Matrix, Code 128, EAN, UPC, PDF417, Aztec, GS1 barcodes, or any 1D/2D barcode type. Also use when someone mentions "DevExpress Barcode", "BarCode", "DevExpress.BarCodes", "generate QR code C#", "barcode image .NET", "create barcode programmatically", or asks about any barcode generation with DevExpress. Covers both .NET and .NET Framework.EngineeringNo warnings★ 67↓ —
- pytorch-lightning-distributedKalarisLabs/research-agent-skillsHigh-level PyTorch framework with Trainer class, automatic distributed training (DDP/FSDP/DeepSpeed), callbacks system, and minimal boilerplate. Scales from laptop to supercomputer with same code. Use when you want clean training loops with built-in best practices.EngineeringNo warnings★ 79↓ —
- devexpress-office-file-api-ai-powered-extensionsDevExpress/agent-skillsBuild .NET applications with the DevExpress AI-Powered Extensions for Office File API to add NLP-powered document processing capabilities — proofreading, translation, and text transformation for Word documents, PDF files, and PowerPoint presentations. Use when integrating AI language models (Azure OpenAI, OpenAI, Ollama, Google Gemini) with document processing workflows, translating documents programmatically, or proofreading documents with AI. Also use when someone mentions "DevExpress AI", "AIIntegration.Docs", "AIDocProcessingService", "proofread Word document", "translate PDF", "AI document processing .NET", or asks about AI-powered document automation with DevExpress. Requires DevExpress v25.2+.EngineeringNo warnings★ 67↓ —
- fine-tuning-serving-openpiKalarisLabs/research-agent-skillsFine-tune and serve Physical Intelligence OpenPI models (pi0, pi0-fast, pi0.5) using JAX or PyTorch backends for robot policy inference across ALOHA, DROID, and LIBERO environments. Use when adapting pi0 models to custom datasets, converting JAX checkpoints to PyTorch, running policy inference servers, or debugging norm stats and GPU memory issues.EngineeringNo warnings★ 79↓ —
- ase-repo-diffrse/aseSummarize the currently staged Git changes as a human-readable, intent-grouped narrative. Use when the user wants a concise and brief report of what changed and why. Optionally can check the diff for intent coherence and show a risk and blast radius report.EngineeringNo warnings★ 60↓ —
- ase-code-lintrse/aseLint source code for potential code quality problems. Use when the user wants to "lint" or "check" source code.EngineeringNo warnings★ 60↓ —
- ase-code-explainrse/aseExplains code with WHAT, WHY, ANALOGY, DIAGRAM, CRUXES, and GOTCHAS. Use when you want to know how code works or when the user asks "how does this work?"EngineeringNo warnings★ 60↓ —
- ase-code-insightrse/aseGive insights into the source code.EngineeringNo warnings★ 60↓ —
- ase-code-analyzerse/aseAnalyze the source code for problems in either the logic and semantics and its related control flow, performance and efficiency, or security.EngineeringNo warnings★ 60↓ —
- ase-repo-dissectrse/aseDissect the current Git change set, treated as an epic, domain-wise and logically into cohesive parts and materialize each part in its own dedicated Git WorkTree. Use when the user calls to "dissect", "split", "break up", or "decompose" a large change set into atomic, separately committable parts.EngineeringNo warnings★ 60↓ —
- ase-repo-resolverse/aseResolve the Git merge conflicts of a working directory -- of an in-progress merge, rebase, cherry-pick, revert, stash apply, or of conflict markers left by patch -- carefully and semantically, never losing any change, escalate unresolvable hunks as-is, continue the in-progress operation, and emit a RESOLVED, PARTIAL, NONE, or FAILED verdict. Use when the user calls to "resolve" or "fix" merge conflicts.EngineeringNo warnings★ 60↓ —
- update-docsDasDigitaleMomentum/opencode-processing-skillsUpdate existing project documentation after code changes. Detects outdated sections, updates module and feature docs, and ensures cross-references remain valid. Use this skill after implementing features, refactoring, or when documentation is known to be stale.EngineeringNo warnings★ 61↓ —