Agent skills that run markdown, that ask for Bash, in other format
A skill is a Markdown file that tells a coding agent how to do one job. We parse every one we can find — the frontmatter, the outline, the languages in its code blocks — so you can look for the one that fits instead of reading through a repository to find out.
69,865 skills from 9,574 repositories, last read 11 September 2026.
Skills declare what they need inconsistently: 58.8% name the languages they run, 11.2% list the tools they ask for, and 4.6% give themselves a category. Filtering narrows to the ones that said so, never to the ones that are.
- gan-style-harnesssifxprime/kodelyth-ecc
GAN-inspired Generator-Evaluator agent harness for building high-quality applications autonomously. Based on Anthropic's March 2026 harness design paper.
otherbashmarkdownReadWriteEdit - eval-harnesssifxprime/kodelyth-ecc
Formal evaluation framework for Claude Code sessions implementing eval-driven development (EDD) principles
otherbashmarkdownReadWriteEdit - existing-repoalinaqi/maggy
Analyze existing repositories, maintain structure, setup guardrails and best practices
otherbashjavascriptjsonmarkdownReadGlobGrep - code-reviewalinaqi/maggy
Mandatory code reviews via /code-review before commits and deploys
otherbashmarkdowntomlyamlReadGlobGrep+1 files - securityalinaqi/maggy
OWASP security patterns, secrets management, security testing
otherbashgitignorejsonmarkdownReadGlobGrep - blender-motion-state-inspectiontempletongroup/radiant
Use this skill when inspecting Blender characters, rigs, poses, animation retargeting, ground contact, facing direction, or model-vs-motion alignment where screenshots alone are no
othermarkdownReadWriteEdit - gan-style-harnesstempletongroup/radiant
GAN-inspired Generator-Evaluator agent harness for building high-quality applications autonomously. Based on Anthropic's March 2026 harness design paper. Use when a feature should
otherbashmarkdownReadWriteEdit - eval-harnesstempletongroup/radiant
Formal evaluation framework for Claude Code sessions implementing eval-driven development (EDD) principles. Use when a Claude Code workflow needs a formal eval before it is trusted
otherbashmarkdownReadWriteEdit - wiki-growflonat/flonat-research
Auto-promote eligible project knowledge articles into Research Vault concepts with generated provenance metadata. Use when running the scheduled knowledge-corpus promotion or previ
othermarkdownBashRead - master-skillswaylq/master-skill
大师.skill — 输入「我做的细分行业」，自动完成行业大佬调研 / 工具地图 / 工作流 / 知识正典 / 信息源 / 术语标准 六轨深度调研，提炼为可运行的「行业 Master OS」skill。 生成的 {industry}.master skill 是自包含目录，所有 hermit agent / Claude Code agent / OpenC
otherbashmarkdownReadWriteEdit+2 files - planning-with-filesjd-opensource/JoySafeter
Implements Manus-style file-based planning for complex tasks. Creates task_plan.md, findings.md, and progress.md. Use when starting complex multi-step tasks, research projects, or
otherbashmarkdownReadWriteEdit+5 files - literature-reviewbeita6969/ScienceClaw
Conduct comprehensive, systematic literature reviews using multiple academic databases (PubMed, arXiv, bioRxiv, Semantic Scholar, etc.). This skill should be used when conducting s
otherbashmarkdownpythonReadWriteEdit - loom-code-reviewcosmix/loom
Comprehensive code review covering correctness, maintainability, performance, security, and best practices.
othermarkdownpythonReadGrepGlob - loom-threat-modelcosmix/loom
Threat modeling methodologies (STRIDE, DREAD, PASTA, attack trees) for secure architecture design.
othermarkdownReadGrepGlob - loom-plan-writercosmix/loom
REQUIRED skill for creating Loom execution plans.
othermarkdownmermaidyamlReadGrepGlob - plannerVinix24/vnx-orchestration
Planning specialist that breaks a VNX feature or track into small, independently-shippable PRs with dependencies and per-PR task-class routing. USE THIS when the user wants to brea
othermarkdownReadWriteEdit - dreamwrg32786/aigent-os
Evidence-gated synthesis over recent operating evidence. Appends admitted lessons and a dated synthesis with propose-only improvement candidates.
othermarkdownReadWriteBash - sweep-nowwrg32786/aigent-os
Run the three read-mostly vault sweeps for stale tracker items, dormant-note candidates, and broken wikilinks, then append one dated result to SWEEP_LOG.md.
othermarkdownReadEditBash - reconcilewrg32786/aigent-os
Read-only attention reconciliation. Computes actual seven-day project attention from daily notes against explicit Tier-1 intended shares.
othermarkdownReadBash - knoxhub-blogrediumvex/seo-blog-writer-claude
Transform any research material into SEO+GEO-optimized blog posts for Knox Hub. Takes any input (URL, article text, notes, video transcript, competitor post, topic idea) and rewrit
otherhtmlmarkdownBashReadWrite - create-bossnicepkg/boss-skill
Distill your boss into an AI Skill — PUA detection (10 corp flavors), counterattack coach, evidence collector, labor law lookup, cake-promise BS meter, resignation script, replacem
otherbashjsonmarkdownReadWriteEdit - self-improving-agentzhouguoqing/QianYuan.AIAgenticFramework
A universal self-improving agent that learns from ALL skill experiences. Uses multi-memory architecture (semantic + episodic + working) to continuously evolve the codebase. Auto-tr
otherjsonmarkdownyamlReadWriteEdit - eval-harnessJamkris/everything-gemini-code
Formal evaluation framework for Gemini CLI sessions implementing eval-driven development (EDD) principles
otherbashmarkdownReadWriteEdit - gan-style-harnessJamkris/everything-gemini-code
GAN-inspired Generator-Evaluator agent harness for building high-quality applications autonomously. Based on Anthropic's March 2026 harness design paper.
otherbashmarkdownReadWriteEdit - ag-destilarandregusman-raiz/a-gusman-claude
Comprime documentos grandes para formato LLM-optimal. Mantem toda informacao em menos tokens. Para TOTVS KB, Design Library, SPECs grandes, docs de referencia. Inspirado no BMAD di
othermarkdownReadGlobGrep - ag-adversarioandregusman-raiz/a-gusman-claude
Adversarial review cross-model: tenta quebrar o design antes do build. Ato 1 = 5 lentes (input/timing/deps/estado/suposicoes). Ato 2 = juiz independente OpenAI Codex (read-only) em
othermarkdownReadGlobGrep - ag-1-construirandregusman-raiz/a-gusman-claude
Maquina autonoma de construcao. Feature, refactor, UI, issue, integracao, otimizacao — recebe objetivo, entrega PR pronto. Padrao MERIDIAN: fases, convergencia, state, self-healing
otherbashmarkdownReadWriteEdit - ag-mesa-redondaandregusman-raiz/a-gusman-claude
Debate multi-agente: 2-4 perspectivas (PM, Arquiteto, QA, Security) debatem decisao tecnica. Output: decisao + rationale + dissenting opinions → ADR. Inspirado no Party Mode do BMA
othermarkdownReadGlobGrep - ag-avaliar-ux-design-libraryandregusman-raiz/a-gusman-claude
Avalia qualidade UX de um projeto comparando screenshots reais contra a Design Library (24 solucoes + design system). Captura telas via Playwright, pontua aderencia a tokens/patter
othermarkdownReadWriteEdit - ag-auditar-react-hooksandregusman-raiz/a-gusman-claude
Auditor de Hooks React. Detecta violacoes das Rules of Hooks, deps faltando/excessivas em useEffect/useMemo/useCallback, candidatos a custom hook, useState que deveria virar useRed
otherbashmarkdowntypescriptReadWriteEdit - ag-curador-skillsandregusman-raiz/a-gusman-claude
Curador trimestral de skills externas (Anthropic + comunidade). Avalia o que pegou no ecossistema, compara com sistema ag local, e recomenda absorver/wrapper/ignorar. Nao instala a
otherbashjsonmarkdownReadWriteBash - ag-melhorar-agentesandregusman-raiz/a-gusman-claude
Analisa reports dos outros agentes, identifica padrões de falha, propõe melhorias nos prompts. O meta-agente anti-frágil.
othermarkdownReadWriteEdit - ralpheddiebelaval/squire
Autonomous AI building loops for long-running projects. Use /ralph <task> for hybrid mode (checkpoints) or /ralph --mode=auto <task> for full autonomy. Designed for overnight build
otherbashmarkdownyamlReadWriteEdit - swarm-iosmrokoss21/swarm-iosm
Orchestrate complex development with AUTOMATIC parallel subagent execution, continuous dispatch scheduling, dependency analysis, file conflict detection, and IOSM quality gates. An
otherbashmarkdownpythonReadGrepGlob+7 files - repo-scanhaibindev/repo-scan
对指定项目源码目录执行全面资产审计，生成《全网模块与源码资产审计详细清单》。当用户要求"审计源码"、"盘点代码资产"、"生成清单"时自动触发。
otherbashjsonmarkdownBashReadGlob+1 files - adraaronsb/claude-code-config
Manage Architecture Decision Records using the project's ADR CLI tool. Use when the user wants to create, list, view, lint, or index ADRs, or when working with docs/architecture/ f
otherbashmarkdownBashReadGrep - scientific-paper-writereniktab/scientific-paper-writer
Full-lifecycle scientific paper writing assistant for any journal. Invoke for: drafting/revising manuscripts, structuring arguments, statistical reporting, citation verification, p
otherbibtexmarkdownyamlReadWriteEdit - plan-and-buildjattanjie21/plan-and-build
Use when a user wants to plan and implement a new feature or bug fix from scratch. Triggers on requests like "I want to build X", "add a feature for Y", "fix Z", or when the user i
otherdotmarkdownReadGrepGlob - planning-with-filesjackjin1997/ClawForge
Implements Manus-style file-based planning for complex tasks. Creates task_plan.md, findings.md, and progress.md. Use when starting complex multi-step tasks, research projects, or
otherbashmarkdownpowershellReadWriteEdit+5 files - evermem-sync-contextnanxingw/EverMem
Query EverMemOS for this project's development history, current state, and future roadmap, then write the result into the current AI tool's context file (CLAUDE.md, .cursorrules, A
otherbashmarkdownBashReadWrite - vsynthphamcuong21478/rtl-skills
Synthesize a Verilog module or full IP with Vivado and report resource utilization, Fmax, and power
otherbashmarkdowntclReadWriteBash - self-improvementbokan/claude-skill-self-improvement
Analyze conversation history to find friction patterns and suggest CLAUDE.md/skill improvements. Use when user wants to review what went wrong across sessions and systematically im
otherbashmarkdownReadBashGrep
Browse all 69,865 skills →200 at a time, grouped by owner. The list above is the skills with the highest standing; this is every one of them.
The whole corpus is one call away through the search API, or ask a question of it on the index itself.