Context & AI engineering
25,729 rows, by stacks then stars
1,865 of these skill files have been read, by 1,328 distinct authors, and what they tell an agent counted. What Context & AI engineering authors agree on, and what they forbid
lilabrooks/lila-agent-skills/skills/write-repository-readme Skill
1★ repoCreate, rewrite, audit, and polish repository README.md files for new or existing projects with Codex, Claude Code, or another Agent Skills-compatible host. Use when asked to draft a README, assess its structure or professional presentation, improve onboarding and scanning, add or repair a table of contents, verify Markdown links and documented commands, reduce README length by linking repository docs, or reconcile README claims with code, configuration, CI, licenses, and project status.
Deadlymind/nanolama/skills/ai-integration Skill
1★ repoWires Anthropic Claude into a Django/DRF backend — the agentic tool-use loop (model proposes tool_use, app runs it, returns tool_result, repeat), a write-tool safety contract for data-mutating tools, capability limits against prompt injection, an OCR/structured-extraction pipeline validated against a schema before persist, Tavily search, and per-tenant metering. Use when adding a Claude agent or tool loop, exposing a tool that mutates data, doing document OCR extraction, calling the anthropic SDK, choosing a model id, or capping AI spend per tenant. Not for regression-testing agent behaviour, golden sets or LLM-as-judge scoring (see ai-evals), the Celery worker that runs a long call (see celery-tasks), or auditing a tool's blast radius (see security-review).
Deadlymind/nanolama/skills/skill-authoring Skill
1★ repoAuthors or fixes a skill in the nanolama library — the SKILL.md frontmatter contract (directory equals name, [a-z0-9-], no reserved words), a third-person description carrying "Use when" and a "Not for" boundary with no colon-space, the five required body sections, one inline example, and backticked See-also names from the roster. Use when creating a new skill, copying template/, editing an existing SKILL.md, wiring scripts/catalog.yaml, or fixing a validate_skills.py failure. Not for authoring the domain content of a specific skill (that lives in the skill itself) or reviewing a code diff (see code-review).
alidevai/chiron_agent/.claude/skills/experience-memory Skill
no license1★ repoToken-verimli kendi kendine ogrenme. Goreve baslarken gecmis dersleri leksikal olarak geri cagirir (yalnizca en ilgili 3-5 kisa ders enjekte edilir); gorev bitince genellenebilir, tekrar eden bir ders cikarsa kaydeder. Amac, transkriptleri context'e doldurmadan deneyimden ogrenmektir.
alidevai/chiron_agent/.claude/skills/mcp-security-review Skill
no license1★ repoBir MCP sunucusunu veya tool'u sisteme baglamadan once guvenlik acisindan denetleme prosedurudur. MCP sunuculari dosya sistemi, terminal, e-posta, GitHub, veritabani veya broker gibi yuksek yetkili kaynaklara erisebildigi icin baglanti oncesi zorunlu incelemedir.
alidevai/chiron_agent/.claude/skills/skill-creator-safe Skill
no license1★ repoAranan skill guvenilir kaynaklarda bulunamazsa, birincil kaynaklardan (resmi dokumantasyon, standartlar, akademik makaleler) yeni bir skill uretme prosedurudur. Uretilen skill her zaman experimental baslar ve normal edinme akisindaki tarama/eval/onay kapilarindan gecer.
yigityildiz0/universal-ai-skill-library/skills/archive-variants/subagent-driven-development Skill
no license1★ repoExecute an implementation plan through bounded implementer and independent reviewer agents with disjoint write ownership, explicit handoffs, evidence gates.
yigityildiz0/universal-ai-skill-library/skills/common/agent-access-policy Skill
no license1★ repoConfigure task-scoped file access guidance for AI coding agents using Codex sandbox/approval workflow and explicit task-scoped instructions. Provides.
yigityildiz0/universal-ai-skill-library/skills/common/ai-agent-development Skill
no license1★ repoDesign, implement, and evaluate tool-using AI agents with explicit authority, typed tools, bounded loops, state, observability, cost controls, and failure.
yigityildiz0/universal-ai-skill-library/skills/common/ai-agent-governance Skill
no license1★ repoImplement the 4 Pillars Framework for AI agent governance (Lifecycle, Risk, Security, Observability). Use when deploying autonomous AI agents, implementing.
yigityildiz0/universal-ai-skill-library/skills/common/claude-agent-sdk Skill
no license1★ repoBuild and maintain applications with the current Claude Agent SDK using installed-version-first documentation, bounded tools, permissions, sessions, MCP.
yigityildiz0/universal-ai-skill-library/skills/common/code-semantic-search Skill
no license1★ repoDesign, evaluate, or operate privacy-aware semantic and hybrid search over a large codebase using structural chunks, lexical retrieval, optional embeddings.
yigityildiz0/universal-ai-skill-library/skills/common/comfyui-prompt-engineering Skill
no license1★ repoDiagnose and improve ComfyUI text conditioning with architecture-aware prompts, encoder and node compatibility checks, controlled experiments, and.
yigityildiz0/universal-ai-skill-library/skills/common/context-engineering Skill
no license1★ repoOptimizes agent context setup. Use when starting a new session, when agent output quality degrades, when switching between tasks, or when you need to configure rules files and context for a project.
yigityildiz0/universal-ai-skill-library/skills/common/cross-model-orchestrator Skill
no license1★ repoCoordinate two or more available AI assistants or model sessions through explicit roles, artifact handoffs, and quality gates. Use for high-risk.
yigityildiz0/universal-ai-skill-library/skills/common/design-handoff Skill
no license1★ repoCreate or review a design-to-engineering handoff with intent, responsive behavior, tokens, states, interaction rules, edge cases, assets, acceptance criteria, and open decisions. Use for design handoff, developer handoff, Figma handoff, implementation spec, or UI acceptance criteria.
yigityildiz0/universal-ai-skill-library/skills/common/dispatching-parallel-agents Skill
no license1★ repoPartition and coordinate two or more independent workstreams without duplicating work, leaking expected conclusions, or creating shared-file conflicts. Use for parallel research, multi-area diagnosis, independent reviews, comparisons, test investigation, or large tasks that benefit from bounded concurrent or batched execution.
yigityildiz0/universal-ai-skill-library/skills/common/figma Skill
no license1★ repoUse the Figma MCP server to fetch design context, screenshots, variables, and assets from Figma, and to translate Figma nodes into production code. Trigger.
yigityildiz0/universal-ai-skill-library/skills/common/gpt-image-2 Skill
no license1★ repoGenerate images with GPT Image 2 (ChatGPT Images 2.0) inside Claude Code, using your existing ChatGPT Plus or Pro subscription — no separate OpenAI access.
yigityildiz0/universal-ai-skill-library/skills/common/gpt-image-edit Skill
no license1★ repoEdit images with OpenAI GPT Image 2 (the `/edit` endpoint of ChatGPT Images 2.0) on RunComfy — bundled with the model's documented prompting patterns so the.
yigityildiz0/universal-ai-skill-library/skills/common/knowledge-base Skill
no license1★ repoCreate, organize, audit, or maintain a durable Markdown knowledge base with an index, atomic notes, stable identifiers, links, tags, and maintenance checks. Use for a personal or team knowledge base, research vault, documentation library, MOC, note organization, broken links, or knowledge-base cleanup; do not confuse it with short-term agent memory.
The same file is in 1 other repository
yigityildiz0/universal-ai-skill-library/skills/common/llm-prompt-engineering Skill
no license1★ repoDesign, harden, evaluate, and version prompts for model-agnostic LLM applications. Use for system instructions, few-shot examples, structured outputs, tool.
yigityildiz0/universal-ai-skill-library/skills/common/mcp-builder Skill
no license1★ repoGuide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK).
The same file is in 42 other repositories
yigityildiz0/universal-ai-skill-library/skills/common/mcp-server-patterns Skill
no license1★ repoBuild MCP servers with Node/TypeScript SDK — tools, resources, prompts, Zod validation, stdio vs Streamable HTTP. Use Context7 or official MCP docs for.
yigityildiz0/universal-ai-skill-library/skills/common/multi-agent-coordinator Skill
no license1★ repoDecompose a complex task into safe parallel and sequential agent work, assign disjoint ownership, track dependencies, reconcile outputs, and verify the.
yigityildiz0/universal-ai-skill-library/skills/common/nimble-agent-builder Skill
no license1★ repoA building experience: create, test, validate, refine, and publish extraction workflows based on existing or new Nimble agents. For users who want to invest.
yigityildiz0/universal-ai-skill-library/skills/common/notebooklm-prompt-architect Skill
no license1★ repoCreate high-quality, source-grounded NotebookLM prompts. Use when the user asks for NotebookLM, Notebook LM, notebook eleme/elemi/kalem prompts, chat.
yigityildiz0/universal-ai-skill-library/skills/common/plan-before-code Skill
no license1★ repoGuide exploration and planning phases before implementation using agentic coding best practices. Includes LLM task suitability assessment, token cost.
yigityildiz0/universal-ai-skill-library/skills/common/prompt-token-optimization Skill
no license1★ repoReduce prompt, tool, retrieval, and conversation token overhead without weakening measured task quality. Use for context-budget audits, oversized.
yigityildiz0/universal-ai-skill-library/skills/common/rag-implementation Skill
no license1★ repoDesign, implement, and evaluate retrieval-augmented generation with traceable ingestion, privacy-aware indexing, hybrid retrieval, grounded answers, and.
yigityildiz0/universal-ai-skill-library/skills/common/subagent-driven-development Skill
no license1★ repoExecute an implementation plan through bounded implementer and independent reviewer agents with disjoint write ownership, explicit handoffs, evidence gates, and integration tests. Use when tasks are separable or independent review materially reduces risk; use the current host's native delegation without model/provider switching.
yigityildiz0/universal-ai-skill-library/skills/common/task-coordinator Skill
no license1★ repoCoordinate complex multi-step tasks by breaking them down into manageable subtasks with dependency tracking. Covers multi-agent architectural patterns (supervisor, swarm, hierarchical), token economics, and handoff protocols. Use when implementing large features, coordinating parallel work streams, designing multi-agent systems, or managing complex workflows.
yigityildiz0/universal-ai-skill-library/skills/common/temporal-orchestration Skill
no license1★ repoDesign, implement, and verify durable Temporal workflows and activities for long-running or failure-prone processes. Use for retries, timers, signals.
yigityildiz0/universal-ai-skill-library/skills/common/ui-component-generation Skill
no license1★ repoGenerate framework-specific UI components by prompting the agent's own LLM directly - no external generation service or MCP required
yigityildiz0/universal-ai-skill-library/skills/common/workflow-orchestrator Skill
no license1★ repoDesign and execute end-to-end workflows by chaining multiple skills and managing quality gates between phases. Use for complete feature implementations.
yigityildiz0/universal-ai-skill-library/skills/platforms/codex/chatgpt-apps Skill
no license1★ repoBuild, scaffold, refactor, and troubleshoot ChatGPT Apps SDK applications that combine an MCP server and widget UI. Use when Codex needs to design tools.
yigityildiz0/universal-ai-skill-library/skills/platforms/codex/codex-agent-dev-kit Skill
no license1★ repoConvert Claude Code agent-kit setups into Codex-compatible skills, project instructions, guardrail scripts, command workflows, subagent guidance.
yigityildiz0/scientific-agent-skills/skills/common/knowledge-base Skill
no license1★ repoCreate, organize, audit, or maintain a durable Markdown knowledge base with an index, atomic notes, stable identifiers, links, tags, and maintenance checks. Use for a personal or team knowledge base, research vault, documentation library, MOC, note organization, broken links, or knowledge-base cleanup; do not confuse it with short-term agent memory.
The same file is in 1 other repository
T0mSIlver/skills/delegate-to-codex Skill
1★ repoCall the OpenAI Codex CLI (`codex exec`) non-interactively to get a second opinion, run a code review, or delegate read-only or edit-capable work to an independent Codex run. Use for Codex CLI subagent-style delegation, long-running worker runs, worktree-isolated edits, machine-readable JSONL output, resume flows, and harness gotchas around stdin and sandboxing.
T0mSIlver/skills/delegate-to-opencode Skill
1★ repoCall the opencode CLI (`opencode run`) non-interactively to get a second opinion, run a code review, or delegate read-only or edit-capable work to an independent opencode run. Use for GLM-5.2/opencode harness delegation, primary/all agent configs with edit permissions, worktree-isolated edit workers, JSON event capture, session resume/fork flows, and gotchas around `mode: subagent`, `--auto`, prompt files, and `--dir`.
M0rtalPhe0nix/dotfiles/.agents/skills/handoff Skill
1★ repoCompact the current conversation into a handoff document for another agent to pick up.
The same file is in 11 other repositories
M0rtalPhe0nix/dotfiles/.claude/skills/claude-md-improver Skill
1★ repoAudit and improve CLAUDE.md project instructions. Use when the user asks to check, audit, update, improve, or maintain CLAUDE.md files or project memory.
M0rtalPhe0nix/dotfiles/.claude/skills/review-agent Skill
1★ repoPerform a read-only, defect-first review of a specified code change and return every actionable finding. Use when another agent delegates review of uncommitted changes, a base-branch diff, a commit, or custom review instructions.
M0rtalPhe0nix/dotfiles/.claude/skills/skill-creator Skill
1★ repoGuide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Codex's capabilities with specialized knowledge, workflows, or tool integrations.
The same file is in 3 other repositories
ychampion/morph-public/skills/morph-install Skill
1★ repoInstall, configure, update, or verify Morph for Codex, Claude Code, Hermes Agent, OpenCode, or OpenClaw from the morph-public repository. Use when the user asks to set up Morph or connect its skill and MCP server. Do not use for creating or editing product stories.
jpbaking/agentic-tests/skills/shared/agentic-refactor Skill
1★ repoRefactor or optimize main code while the agent-test suite (and user tests) must stay green after every change. The inverse of agentic-unit-test — main code may change, tests may not. Use when asked to refactor, optimize, or clean up code that agent tests already lock down.
jpbaking/agentic-tests/skills/shared/agentic-test-clean Skill
1★ repoRemove every agent-generated test file (*.agentic.spec.*, *.agentic.test.*, *AgenticTest.java, test_agentic_*.py, *_agentic_test.c/.cpp) plus the agentic plan files, after an explicit user confirmation. Use when asked to delete, clean up, or remove the agent tests or start the agentic-test lifecycle over. Never touches main code, user tests, or test config.
jpbaking/agentic-tests/skills/shared/agentic-test-update Skill
1★ repoReconcile agent tests after the user INTENTIONALLY changed main-code behavior — show each failing agent test as old-locked vs new-actual behavior, and update tests only with per-diff user confirmation. Use when agent tests fail after a deliberate behavior change.
jpbaking/agentic-tests/skills/shared/agentic-unit-test Skill
1★ repoGenerate agent-authored unit tests ("agent tests") to reach a user-defined coverage goal for one target file/class or the whole project, without ever modifying main code or user-made tests. Use when asked to write unit tests, raise coverage, or lock down existing behavior before refactoring/optimization.
palol/skills-zoo/skills/dg-skill-authoring Skill
1★ repoAuthor a new Agent Skill that adds a feature to an Obsidian Digital Garden (oleeskild digital-garden plugin + Eleventy) site. Use when a user wants to package a Digital Garden customization - a component, widget, layout tweak, or style - as a reusable, upstream-safe SKILL.md that another DG user's agent can install. Produces a spec-compliant skill (frontmatter, progressive disclosure, references, assets), verifies it with skills-ref validate, and follows the user-owned-paths convention so upstream git pull never clobbers it. Trigger on: make a digital garden skill, package this DG customization, write a skill for my digital garden, meta-skill, skill authoring, turn this into a skill.
smkuxi71-hash/skill-ship-max/core/.claude/skills/ship-claude-max-orchestrate Skill
no license1★ repoImplementation-only release-slice G3 loop для ship-claude-max: Claude пишет код по TDD, GPT-5.5/gpt-5.6-sol medium держит adversarial Coach, hard budgets, build leases и committed review ledger.
smkuxi71-hash/skill-ship-max/core/.claude/skills/ship-claude-max Skill
no license1★ repoValue-gated, release-sliced design-to-production pipeline с Claude-исполнителем: Claude Code — Conductor и единственный writer, GPT-5.6-sol — со-конструктор дизайна, GPT-5.5/medium — plan/code review и Coach, exact-SHA release.
Paldom/skillskit/skills/create-skill-repo/assets/template/.claude/skills/add-skill Skill
1★ repoAuthors or improves an Agent Skill in this repo via the eval-first workflow - scope one purpose, write trigger evals, draft SKILL.md, validate, iterate. Use when the user asks to create, add, write, refine, or fix a skill ("add a skill for X", "improve triggering of Y"). Not for scaffolding a new repository or editing infrastructure (hooks, CI, the validator/scripts, docs).
LeonardSEO/power-platform-skills-codex/plugins/canvas-apps/skills/configure-canvas-mcp Skill
1★ repoConfigure the Canvas Authoring MCP server for the current coauthoring session. USE WHEN "configure MCP", "set up MCP server", "MCP not working", "connect Canvas Apps MCP", "canvas-authoring not available", "MCP not configured", "set up canvas apps".
The same file is in 1 other repository
LeonardSEO/power-platform-skills-codex/plugins/code-apps-preview/skills/add-mcscopilot Skill
1★ repoAdds Microsoft Copilot Studio connector to a Power Apps code app. Use when invoking Copilot Studio agents, sending prompts to agents, or integrating agent responses.
The same file is in 1 other repository
LeonardSEO/power-platform-skills-codex/plugins/mcp-apps/skills/generate-mcp-app-ui Skill
1★ repoGenerate an MCP App widget (self-contained HTML) for an MCP tool. Describe the visual you want and paste your tool's test output. Use when user asks to create an MCP App, widget, or visual for a tool.
The same file is in 1 other repository
jahfaliabdulrahman-dev/hermes-skills/skills/flutter-soul-stewardship Skill
1★ repoProduction-grade agent skills for Flutter, DevOps & AI Governance
jahfaliabdulrahman-dev/hermes-skills/skills/swarm-executive-controller Skill
1★ repoProduction-grade agent skills for Flutter, DevOps & AI Governance
dkmqflx/git-workflow-plugin/plugins/git-workflow/skills/worktree-done Skill
1★ repoRemove the current git worktree, switch the main checkout to the base branch, and delete the feature branch. Use when the user has finished work started with the worktree skill (or `/worktree`) and wants to clean up — phrases like "워크트리 정리해줘", "워크트리 삭제해줘", "clean up this worktree", "remove worktree", "/worktree-done". Do NOT use from the main checkout (not a worktree) — that is the branch-done skill.
The same file is in 1 other repository
dkmqflx/git-workflow-plugin/plugins/git-workflow/skills/worktree Skill
1★ repoCreate a new git worktree on a fresh branch and switch into it before starting work. Use when the user asks to start a task in a new worktree, work in isolation from the main checkout, or invokes "/worktree" — phrases like "워크트리 만들어줘", "새 브랜치 파서 작업해줘", "worktree에서 작업 시작". Do NOT use for removing or cleaning up an existing worktree.
The same file is in 1 other repository