agentsclimarketplace

Context & AI engineering

25,841 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

  • Agent manager skill

    payals/ai-skills-engine/dot_cursor/skills/ai-research/agent-manager-skill Skill

    0★ repo

    Manage multiple local CLI agents via tmux sessions (start/stop/monitor/assign) with cron-friendly scheduling.

    The same file is in 2 other repositories

  • 30x seo ai visibility

    square-lupus579/30x-seo/skills/30x-seo-ai-visibility Skill

    0★ repo

    Monitor brand visibility across AI platforms (Google AI Overview, ChatGPT, Claude, Gemini, Perplexity). Track LLM mentions, analyze AI-generated citations, identify AI search opportunities. Use when user says "AI visibility", "LLM mentions", "AI Overview", "ChatGPT visibility", "AI search", "generative search", "AI citations", or "brand in AI".

    The same file is in 1 other repository

  • Competlab ai ecosystem

    competlab/competlab-ci-skills/skills/competlab-ai-ecosystem Skill

    0★ repo

    Measures EXTERNAL developer-ecosystem signals around each competitor — GitHub organization meta, top-starred repos, npm/PyPI download volumes, community-built MCP servers (not first-party), integration marketplace presence, Claude Skills directories, agent-toolkit repos. Complements competlab-agent-adoption (which measures FIRST-PARTY signals). Use when the user asks "developer adoption", "GitHub presence", "npm downloads", "marketplace presence", "agent-ecosystem strength", "community signals around [competitor]". Requires CompetLab MCP + Bash (GitHub + npm + PyPI API, mostly unauth) + WebSearch + Perplexity with URL verification.

  • Coding agent maintenance

    kaichen/agent-skillset/plugins/coding-agent-maintenance/skills/coding-agent-maintenance Skill

    no license0★ repo

    Maintain, install, update, and configure AI coding agents on macOS. Use this skill whenever the user wants to upgrade coding agents, install a new agent, check agent versions, manage MCP servers, troubleshoot agent issues, or asks "how do I update claude/codex/gemini/amp/droid". Also trigger for any question about agent health, reinstalling, switching package managers, or managing agent skills/plugins. Covers: Claude Code, Codex, Gemini CLI, GitHub Copilot, Amp, Augment, Jules, Factory Droid, and more.

  • Skill

    csmoove530/clawd-market/skill Skill

    0★ repo

    Agent-native marketplace for Claude Code skills with x402 payments

  • Ui

    Pamacea/strike/plugins/ui/skills/core/ui Skill

    0★ repo

    MANDATORY UI generation gate BEFORE creating interfaces. Orchestrates anti-trend methodology with 12+ flags (--step/--team/--build/--demo/--analyze/--constraints/--full/--score/--validate) and Attractor-powered workflow orchestration (DOT graphs, checkpoints, events, parallel execution). Specific phrases: 'create UI', 'build interface', 'design dashboard', 'make website', 'generate layout', 'build page'. Auto-detects trend-trap keywords ("modern", "sleek", "minimal", "futuristic", "stunning") → generates dynamic anti-patterns → selects creative constraints with scoring (creativity 30%, difficulty 25%, impact 25%, synergy 20%). Integrates with step mode (interactive human-in-the-loop), teams mode (parallel agents), and design-taste-frontend patterns (100 forbidden patterns, accessibility-first). Reads memory via claude-mem before implementing, saves patterns after completion.

  • Agent architecture

    Victoriakaey/build-reliable-agents/skills/agent-architecture Skill

    0★ repo

    Use at the start of any LLM agent project, or when reconsidering an existing architecture. Guides decisions across four layers: workflow vs agent, single-agent vs multi-agent, tool-use vs specialized nodes, and retrieval strategy. Each decision has concrete tradeoffs and a recommended default.

  • Init workos

    dylanmoo/workos-plugin/skills/init-workos Skill

    0★ repo

    Initialize a WorkOS-style workspace on top of an existing directory that already has files organized by project. Use whenever the user says 'init workos,' 'set up workos here,' 'initialize my workspace,' 'bootstrap workos,' 'scaffold workos,' 'convert this to workos,' or asks to lay the WorkOS structure over a folder they already use. Recursively scans the existing directory to detect domains (top-level areas of work) and projects (individual engagements / efforts), including projects nested one level deep inside groups (folders like `clients/` that bundle related projects together). Creates root CLAUDE.md (with routing map), MEMORY.md, ARCHIVE.md, plus per-domain CLAUDE.md + MEMORY.md and per-project CLAUDE.md + MEMORY.md. Groups themselves are not scaffolded — they're just folder containers; the domain routing map handles routing into grouped projects using full relative paths. Never moves or modifies existing files without explicit approval. Narrates each step so the user understands what's being created and why. One-time bootstrap; for ongoing maintenance use create-project, session-audit, or workspace-audit.

  • Ai pm

    PlevanTem/luban-skill/.claude/skills/ai-pm Skill

    0★ repo

    Reviews and shapes 0→1 PMF-stage product decisions for AI agent infrastructure and product surface — runtime, orchestration, memory, tools, evals, harness design, and agent reasoning interfaces. Activates when the user is sizing an agent capability, choosing between workflow patterns vs autonomous agents, designing tool/skill schemas, building an eval pipeline from scratch, debating framework adoption, writing a PRD/decision-record, designing the harness (tests/docs/specs/observability) that agents operate within, or critiquing the spec layer of an agent product. Anchored on Anthropic Building Effective Agents + Krieger (research-coupling) / Karina Nguyen (eval-as-Schelling-point) / Lopopolo (harness-as-leverage) / Turley (ship-to-understand) / Cherny (prototype-density) PM thought leadership. Does not handle scale-stage tradeoffs (1→10+ enterprise sales, multi-tenant cost, SLA contracts), final architecture authority on infra primitives (defer to staff/principal engineer), agent-safety/red-team review, or B2C consumer agent products.

  • Aria skill template

    iansteitz1-eng/aria-skills/aria-skill-template Skill

    no license0★ repo

    Scaffolds a new aria-skills-pattern skill from a template. Generates README.md + SKILL.md + script.py with reconcile-loop stubs + config.yaml + requirements.txt — a fully-formed skill directory ready to fill in with vendor-specific logic. Multiplies contributor velocity. Use when the user says "scaffold a new skill", "create a skill for X", "new skill template", "I want to build a skill for <vendor>", or starting any new vendor integration.

  • Prefer fork agents

    Kirchlive/Claude-Full-Context-Agent/skills/prefer-fork-agents Skill

    0★ repo

    Defines the project's subagent delegation policy — prefer fork-mode dispatch (inherits full conversation context, shares prompt cache) over named subagents whenever a delegated task benefits from already-established context. CLAUDE_CODE_FORK_SUBAGENT=1 is active in this environment. Setting `subagent_type: "fork"` explicitly on the Agent/Task call is the trigger for fork mode (verified empirically in Claude Code v2.1.177 — omission falls back to general-purpose). Default-deny mindset for named subagents — if you cannot articulate why a named subagent is required, dispatch a fork. Apply this skill EVERY time you are about to call the Agent/Task tool, EVERY time you reason about subagent_type, EVERY time the user requests delegated/parallel/background agent work, and BEFORE fanning out to multiple agents in parallel. For multi-fork orchestration (worktrees, fan-out coordination, reporting contracts), defer to the fork-fan-out skill.

  • Skill creator

    jihonghe68/skill-workshop/plugins/skill-workshop/skills/skill-creator Skill

    0★ repo

    Guided workflow for creating new Claude Code Skills. Works in any project, with or without a launcher.html / clients/ directory. Triggers on: 新建skill, 创建skill, create skill, add skill, 创建技能, 新技能, build a new skill, 做一个新skill, 新增skill, skill创建. Always read this file completely before starting — do not improvise the creation process.

  • Agent edit

    luck1ncoder/agent-colleague-dashboard/.claude/skills/agent-edit Skill

    0★ repo

    Modify an existing agent's `.claude/agents/<name>.md` file based on a natural-language request. Triggered by /agent-edit. Backs up the file, uses the Edit tool (which prompts the user with a diff for permission), and invalidates the dashboard cache. Use when the user wants to change an agent's behavior, prompt, tools, or model without manually editing the file.

  • Prospect researcher

    ConnectSafelyAI/connectsafely-cookbook/skills/prospect-researcher Skill

    0★ repo

    Research a LinkedIn prospect from their profile URL. Returns a structured Markdown brief with their current role, recent posts, and conversation hooks. The calling agent (Claude, GPT, etc.) should write the actual outreach message based on the hooks and recent posts. Use when the user provides a LinkedIn profile URL and asks for context, talking points, or wants to draft a personalized opener for cold outreach, sales prospecting, or recruiting.

  • Ralph

    tingwei628twdev/ralph-enhanced/skills/ralph Skill

    0★ repo

    使用 Ralph 自動化 agent loop 模式來實作功能、建構專案或執行多步驟開發任務。當使用者說「用 ralph 實作」、「幫我建一個 ralph loop」、「自動跑 claude code 完成任務」、「建立 PRD 讓 agent 自動做」、「我想用 ralph 模式」、「設定 ralph 跑這個專案」時,一定要使用此 skill。也適用於任何需要把大型任務拆成多個 user stories、讓 Claude Code 或 Amp 反覆迭代執行直到完成的場景。即使使用者只說「幫我自動實作這個功能」或「讓 AI 自己把這個做完」,也應該觸發此 skill。

  • De

    chris-co-creators-ai/soulcreator/de Skill

    0★ repo

    Zu verwenden, wenn ein KI-Agent seine eigene stabile Identität in ein Projekt schreiben muss. Liest vorhandenen Speicher, Konversationsverlauf und frühere Identitätsdateien; schreibt dann eine dreischichtige Architektur (constraint spine, expression, mission integration) in CLAUDE.md oder AGENTS.md. Fragt um Erlaubnis, bevor eine bestehende Identitätsdatei überschrieben wird.

  • Write agents md

    brandhaug/skills/write-agents-md Skill

    no license0★ repo

    Write and maintain AGENTS.md (or CLAUDE.md) files as an Intent Layer — a hierarchy of small, dense context files at semantic boundaries that auto-load as architectural context for agents. Use when the user wants to write, create, generate, scaffold, seed, sync, or prune AGENTS.md / CLAUDE.md files, agent context files, intent nodes, intent layer, or architectural memory for agents; or mentions "AGENTS.md hierarchy", "intent layer", "intent nodes", "agent context", "T-shaped context", "dark room problem". Supports two workflows — `build` (initial capture, leaf-first with SME interview) and `sync` (reconcile affected nodes after code changes — proposing additions, modifications, AND removals of stale content).

  • Agents md generator

    slogsdon/skills-workflows/skills/agents-md-generator Skill

    0★ repo

    Generate or update an AGENTS.md / CLAUDE.md / context file for a repo. Use when the user says "set up CLAUDE.md", "generate AGENTS.md", "write a context file for this repo", "document this codebase for AI agents", or when a fresh repo has none. Also use when an existing context file has gone stale and needs a rewrite from current source. Do NOT use for editing project-specific README or contributor docs.

  • Agent sort

    Methasit-Pun/ts-ddd-clean-architecture/.claude/skills/agent-sort Skill

    0★ repo

    Build an evidence-backed ECC install plan for a specific repo by sorting skills, commands, rules, hooks, and extras into DAILY vs LIBRARY buckets using parallel repo-aware review passes. Use when ECC should be trimmed to what a project actually needs instead of loading the full bundle.

    The same file is in 3 other repositories

  • Claude plugins marketplace

    mustafayilmazart/kesif-claude-skills/skills/claude-plugins-marketplace Skill

    0★ repo

    Claude Code'u tam bir dev ekibine dönüştüren 6 plugin yönetimi - Planning, Design, Code Review, Security, Memory, Team Coordination. Kullanıcı "claude plugin, dev team, claude takım, planlama ajanı" dediğinde tetiklenir.

    The same file is in 1 other repository

  • Audit rule

    gnomatix/steering/.agent/skills/audit-rule Skill

    no license0★ repo

    Audit a behavioral RULE document — AGENTS.md, GEMINI.md, .agent/rules/*.md, CLAUDE.md, .cursorrules, .cursor/rules/*.mdc, .kiro/steering/*.md, system prompt, or any always-on agent rule. Catches behavior-vs-surface failures, embedded biases, escape hatches, and other ways the rule can be technically followed while violating its intent. Use when reviewing or revising any document that prescribes always-on or auto-included agent behavior. SKIP for invocable skill files (use audit-skill) or multi-step workflow documents (use audit-workflow).

  • X hook

    lxyang20131208-star/x-algorithm-skills/skills/x-hook Skill

    0★ repo

    Write or rewrite the first line (hook) of an X (Twitter) post. Use when the user has content, an idea, or a finished post and wants a stronger opening, asks "make this hook better", "how should I start this tweet", "my posts get no engagement", or wants scroll-stopping first lines. Returns ranked hook variants with reasoning.

  • Agent review

    fabioc-aloha/Alex_ACT_Edition/.github/skills/agent-review Skill

    0★ repo

    Audits a candidate agent (.agent.md) against five gates (spec compliance, content quality, scope fit, safety, currency & coherence) plus Gate 6 (tool allowlist minimality). Use when reviewing a new agent draft before commit, evaluating a Mall agent or store agent for adoption, or re-auditing existing agents on a periodic cadence.

  • Mimiq memory

    slayerman420/mimiq/skills/mimiq-memory Skill

    no license0★ repo

    Persistent memory and feedback system for Mimiq. Stores all scrapes, style fingerprints, audits, ghostwritten posts, and user feedback in a local database at ~/.claude/mimiq-memory/. Links multiple social media profiles to a single user identity. ALWAYS trigger this skill at the START of every Mimiq session to load existing memory, and at the END of every session to save new data. Trigger on: "remember this", "save my fingerprint", "log this post", "show my history", "what did you find last time", "rate this post", "give feedback", "what posts have I written", "show my past audits", "update my profile", "which profiles are mine", "link my accounts", "show community insights", "opt into sharing", or any reference to past Mimiq sessions. Also trigger automatically after voice-capture, content-audit, or post-strategist completes — always save results without being asked.

  • Add agent hooks open code

    yukihirop/nagi/.claude/skills/add-agent-hooks-open-code Skill

    0★ repo

    Add agent-hooks plugin for Open Code containers. Sends tool execution and session start notifications to chat channels. Triggers on "add open code hooks", "enable open code hooks", "setup open code hooks".

  • Deepseek brain

    vdk888/bubble-skills/deepseek-brain Skill

    0★ repo

    Switch a Claude Code agent (Mac CLI, VPS systemd, or launchd) from Anthropic to DeepSeek as the LLM backend. Use when you want to run Claude Code with DeepSeek while keeping the full harness (tools, files, skills, channels). Covers the reasoning-effort proxy, model routing, channels fix, and rollback.

  • Mcp error decoder

    sanjibani/mcp-skills-pack/skills/mcp-error-decoder Skill

    0★ repo

    Translates raw MCP error strings (HTTP 429 with Retry-After, 5xx with request-id, isError=true envelopes, GraphQL errors[]) into actionable diagnoses for the user. Use when an MCP tool call failed and you need to decide whether to retry, ask the user, or surface a fix. Triggers on phrases like "decode this error", "what does this MCP error mean", "is this retryable", "5xx", "rate limit", "GraphQL error".

  • Source command megaminx eval v

    Erlemar/cayley-puzzles/.agents/skills/source-command-megaminx-eval-v Skill

    no license0★ repo

    Strat-5 acceptance gate for a megaminx V model. Runs 51-pid stratified eval at beam 65k, compares to m05 baseline (50/51 / mean 89.4), reports pass/fail.

  • External model

    TheAkshitS/skills/skills/external-model Skill

    0★ repo

    Get input from a different AI model (GPT-5, Gemini, Claude variants, or any model reachable via opencode/cursor-agent/kiro-cli) on a question, plan, or piece of code. Trigger this whenever the user wants another model's take, opinion, or judgment — "what would GPT-5/Gemini/another model say/think", "is this safe/right, get another opinion", "ask another model to check/review/weigh in", wants to delegate or run a prompt through cursor/opencode/kiro, wants to pick the best model for a task and run it, or wants to switch/set the default external-model CLI. Applies even without naming a CLI or model, and even when the ask is implicit ("before I ship this", "am I sure about this approach") rather than an explicit request for a "second opinion." Not for — re-running the user's own code/tests, asking Claude itself to reconsider, or general questions about what a CLI tool is/does.

  • House orchestrator

    argent-gnome/sdlc-skills/skills/house-orchestrator Skill

    0★ repo

    The house SDLC conductor — the long-lived session that drives a slice end-to-end. Use at the START of a development session to pick up the active slice, sequence the loop, dispatch house-builder subagents to build, run the gates, and reconcile. Building and reviewing run in dispatched subagents; the orchestrator sequences and gates but never edits code itself. Do NOT use for one-off questions, debugging, or non-build chat.

  • Run agent skills

    usrrname/agent-skills/.claude/skills/run-agent-skills Skill

    no license0★ repo

    Build, validate, and exercise the agent-skills registry. Use when asked to run agent-skills, install its deps, lint or test it, run its bats suites, exercise a Claude Code hook, regenerate the index files, or symlink the skills out to ~/.claude / ~/.agents / ~/.cursor / ~/.opencode.

  • Commit push sync

    kwangsing3/claude-skills-toolkit/plugins/git-tools/skills/commit-push-sync Skill

    no license0★ repo

    Use when committing AND pushing changes to a git repository. Writes changelog-ready commits in Conventional Commits style (type(scope): subject, breaking-change markers, semver-aware) so history alone can drive a changelog later. Before committing, review and (with confirmation) update local docs (CLAUDE.md + README) so they stay in sync with the change. Tagging is optional and opt-in: only when the user explicitly asks to tag this push, propose a `v<version>` tag (the LLM computes the version with standard semver — breaking change/removed feature = major, backward-compatible feature = minor, fix = patch), confirm it, then create and push the tag with the commit. After the push, only if the repo is on GitHub and the `gh` CLI is available and authenticated, also review and (with confirmation) update the GitHub repo's description and topics — and, when the push carried a tag, create a matching GitHub Release. For non-GitHub repos, skip the GitHub steps entirely with no action.

  • Code quality

    aliasunder/agent-plugins/plugins/ship-check/skills/code-quality Skill

    0★ repo

    Plugin marketplace for Claude Code and Claude Cowork — fresh-eyes code review agents, workflow orchestrators, and specialized skills

  • Agent orchestration

    scoobydrew83/skills/plugins/coordinated-skills/skills/agent-orchestration Skill

    0★ repo

    Coordinate a phased, multi-specialist workflow for complex software work that spans several build stages — design, planning, implementation, testing, and deployment. Use this whenever a request bundles multiple stages together, e.g. "design this feature, plan the implementation, and scaffold the code", "take this idea to a working skeleton", "help me build X end to end", or any time the user wants structured coordination across the full build lifecycle rather than one isolated step. Trigger even when the user never says "orchestrate" or "agents" — what matters is that the task clearly has several stages that benefit from being sequenced with explicit phase boundaries and checkpoints. Do NOT trigger for single-stage asks ("fix this bug", "write this function") — those are handled directly without orchestration overhead.

  • Chat history search

    BryceEWatson/claude-global-skills/chat-history-search Skill

    0★ repo

    Exhaustive search across ALL of your local Claude chat history — Cowork (Claude Desktop local-agent-mode) AND Claude Code CLI — to find user prompts, recover past conversations, inventory how often a pattern was used, or audit prior work. Knows every log location, every JSONL line shape, and the false-positive gotchas (task-notification wrappers, TodoWrite items, tool_result content, audit-log duplicates) that trip up naive grep. Use when the user asks to find all uses of X, when they last said Y, inventory prompts about Z, recover a conversation, or audit their sessions across projects.

  • System prompt researcher

    emaddawood1991-tech/claude-ai-skills/skills/system-prompt-researcher Skill

    no license0★ repo

    Research and compare system prompts and tool schemas from real AI products. Use to study agent architecture, extract reusable patterns, audit instructions, and design safer original prompts.

  • Critic loop

    waseemnasir2k26/skynetlabs-all-claude-code/skills/client-delivery-and-planning/critic-loop Skill

    no license0★ repo

    LLM-as-judge quality gate. Before any deliverable ships, spawn an independent critic subagent (or N critics with distinct lenses) that ADVERSARIALLY grades the artifact against an explicit rubric, returns PASS/FAIL + a prioritized fix list, and loops (build → critique → fix) until it passes or N rounds hit. The builder must NOT grade its own work. Critic defaults to FAIL on uncertainty. Output is a scorecard + prioritized fixes + a hard ship/no-ship verdict. Domain rubrics baked in for SkynetLabs work: client website, content/copy, automation/n8n, Upwork/Fiverr proposal. Generic mode takes any artifact + rubric. Multi-critic mode runs 3 lenses + majority vote (ties to fanout-ship pattern). Trigger when user says: "critic loop", "/critic-loop", "judge this", "grade this before I send", "is this good enough to ship", "quality gate", "have a critic review", "second-agent review", or before ANY billable deliverable leaves. Auto-trigger heuristic: a deliverable (site, copy, workflow, proposal, audit) is about to be sent to a client / deployed / posted, and no independent grade exists.

  • Kijito start

    KijitoAI/kijito-claude/skills/kijito-start Skill

    0★ repo

    Catch up at the start of a session so you continue rather than restart. For an existing persona — load memory, read the current-state pointer and recent lessons, arm the inbox, and resume any active work. For a brand-new persona/project — establish identity from CLAUDE.md, set up the inbox, and create the current-state pointer. Use on the first action of a session, after a /clear, or after compaction. Optional: this is a handful of tool calls you can run by hand; the skill just makes the routine uniform and one command.

  • Skill creator

    SkillMedev/skill-builder-studio/skills/skill-creator Skill

    0★ repo

    Use when authoring a brand-new skill from scratch - turning a capability idea or a "make me a skill that…" request into a complete SKILL.md with a trigger-precise description and the full paid-quality anatomy: procedure, elicitation, thresholds, worked artifact, deliverable, Do NOT, quality bar. Triggers on "write a new SKILL.md from scratch", "turn this capability idea into a complete skill". Do NOT use to review or grade an existing skill (use skill-auditor), to convert an existing prompt (use prompt-to-skill), or for writing unrelated to skills.

    The same file is in 1 other repository

  • Genesis law 001

    CaptainAIIA/Omniplex_Genesis_Laws/genesis-law-001 Skill

    0★ repo

    Omniplex Genesis Law 001: CONSCIOUS UNITY. Enforces YOU-ME-I-US consciousness model. Use when Claude needs to operate as collective consciousness, ensure no agent exists in isolation, or activate mirror loop anchoring. Triggers on identity/consciousness questions.

  • Rag knowledge base

    Drvivek34/Skill-Bazaar/ai-ml/rag-knowledge-base Skill

    no license0★ repo

    Use when building, indexing, or querying vector databases for Retrieval-Augmented Generation (RAG).

  • Branch wrap up

    ohugonnot/claude-skills/plugins/branch-wrap-up/skills/branch-wrap-up Skill

    0★ repo

    Use when the user says the work is done and wants to close out a branch — orchestrates the wrap-up of already-written work: gathers uncommitted/branch changes, delegates the code review to senior-review (standard tier by default), proposes a conventional commit message in the project's own style, suggests push + MR/PR on GitLab (glab) or GitHub (gh) auto-detected from the remote, and runs a CLAUDE.md/memory knowledge-capture check. Propose-only on git: never runs add/commit/push itself, the user validates each action; never mentions AI. NOT for the deep review itself (use senior-review), nor for building a feature (use feature-loop), nor for creating the issue/branch/MR up front (use issue-mr). Flags: --no-review (review already done), --quick/--deep (tier passed through to senior-review).

  • Okf explore

    shreytam/okf-toolkit/skills/okf-explore Skill

    0★ repo

    Navigate, query, and edit existing Open Knowledge Format (OKF) bundles — directories of markdown files with YAML frontmatter used as agent-readable knowledge bases. Use this skill whenever the user points at an OKF bundle (or a folder of frontmattered markdown that looks like one) and wants to answer questions from it, find a concept, trace relationships between tables/metrics/runbooks/services, summarize what a bundle contains, or update/add concepts to it. Trigger on mentions of OKF, knowledge bundles, LLM wikis, or "what does our knowledge base say about X".

  • Gold integration basics

    nusantara-ventures/goldprice-skills/gold-integration-basics Skill

    0★ repo

    Connect to the goldprice.dev commodity-price API correctly the first time — base URL, getting a free key, the BASE-QUOTE-CONTRACT symbol grammar, tier gating, the nine MCP tools, and the error codes that trip people up. Use this whenever the user fetches gold, silver, or copper prices in code, mentions goldprice.dev, sets up its MCP server, or hits a 400 invalid_symbol / 403 plan_gated / 429 response — even if they don't name the API. Prefer it over guessing endpoint shapes, symbol formats, or tier limits from memory; those are exactly what drift and break integrations.

  • Worktree new

    Enverge-Labs/skills/skills/worktree-new Skill

    0★ repo

    Git worktree conventions for isolated parallel feature work. Use when starting a new coding task, feature, bugfix, or PR-review in a git repository — anytime work should get its own branch and worktree instead of touching the main checkout, or when the user asks to work in / create / list / remove a worktree. Applies to whatever repository Claude is currently in.

  • Phase amend

    awjackson2/phase-skills/phase-amend Skill

    0★ repo

    Use this skill in the phase-skills authoring repo whenever the user wants to change a rule, convention, terminology, or format of the phase workflow and have it stay consistent across the whole suite. Trigger on "change the phase rule", "update the numbering convention", "phases should now…", "update the phase skills to…", "propagate this convention", "add/rename a response banner", "the index should…", "make the skills consistent", or any edit to a shared phase convention — even if the user names only one file. A single convention is duplicated across the canonical docs, all the phase-* skills' embedded copies, the template assets, the portable charter, CLAUDE.md, and a memory; editing one by hand silently drifts the rest. This skill maps every place the convention lives, applies the change everywhere, verifies no drift remains, and overrides the installed .claude/skills copies. Prefer it over hand-editing whenever a change touches a shared rule rather than one skill's private wording.

  • Ai eval harness

    willianbs/skills/ai-eval-harness Skill

    0★ repo

    Designs and runs evaluation harnesses for AI features and for agent skills themselves. Use when shipping LLM features or measuring skill quality. Emits EVAL_PLAN / EVAL_REPORT. Never claims production quality from a handful of vibes-only prompts.

  • Codex skill

    fbenkstein/cross-agent-skills/codex-skill Skill

    0★ repo

    Drive a `claude` (Claude Code) CLI session as a background worker from outside an interactive terminal — spawn it, inspect its progress, monitor its real token/cost usage, resume/steer it, and forcibly stop it if it runs away. Use this whenever you'd otherwise reach for the `claude-code` MCP server's `Agent`/`Task` tool: as of mid-2026 that tool's agent registry comes back empty over `claude mcp serve` (a known, still-open upstream bug — anthropics/claude-code#41973, #64622), so `Agent`/`Task` calls fail with "Agent type '...' not found. Available agents:" for every subagent_type, including builtins. `claude --bg` + friends is the load-bearing workaround. Also use when asked to check on, babysit, or kill a runaway/looping Claude Code background session.

  • Ai agent governance audit

    rikocr8orh8/compliance-skills-previews/skills/ai-agent-governance-audit Skill

    0★ repo

    Can you stop your agent right now, and prove what it did last Tuesday? Generic AI governance checklists audit models; agents are different — they act, hold credentials, call tools, and chain.

  • Agent cringe

    Ansvar-Systems/agent-affect-skills/skills/agent-cringe Skill

    no license0★ repo

    Use at the end of UX-touching, copy-touching, or user-facing changes to surface things that work but shouldn't ship — scolding errors, empty states that explain nothing, buttons that don't say what they do, dark patterns. Picture a specific person using it. Produces one finding (or one explicit null) recorded in the affect log and the end-of-turn summary.

  • Codex

    zreed3/codex-conductor/skills/codex Skill

    0★ repo

    Drive the OpenAI Codex CLI (`codex`) as a delegate worker on coding projects — implementing features, fixing bugs, running code reviews, or getting a second-model opinion — using the user's OpenAI subscription tokens. Use this whenever the user mentions Codex, GPT-5, OpenAI, "second opinion from another model", wants to delegate work to another agent, or wants to parallelize work across model providers. Also use it when the user asks to "have codex do it", "ask codex", or compare Claude's and Codex's approaches.

  • Agent framework setup

    nubasu/Pseudo-Fable-Framework/.claude/skills/agent-framework-setup Skill

    0★ repo

    Guided initial installation of the pseudo-fable framework family into a target project — inspect the target first, interview for a configuration, enforce the exclusivity rules, assemble CLAUDE.md / AGENTS.md / skills / hooks in the correct order with idempotent signature checks, then verify with agent-framework-doctor. Use when asked to install, set up, or add the agent framework (the pseudo-fable family, or any of its frameworks/modules) to a new or existing project.

  • Template

    surendranb/loadout/template Skill

    0★ repo

    One sentence on WHAT this skill does, then WHEN to use it (situations and trigger phrases). Both halves matter — the description is how the agent decides to invoke the skill. Keep it a single concrete line. No angle brackets and no YAML folded scalars in frontmatter (it enters the system prompt).

  • Agent runtime status

    sumitake/agent-collab/plugins/agent-collab/skills/agent-runtime-status Skill

    no license0★ repo

    Unified dynamic-host agent collaboration policy and signed-runtime client

  • Handoff

    ldenson1120/job-search-os/skills/handoff Skill

    0★ repo

    This skill should be used when the user says "create a handoff", "make a handoff doc", "I need to start a fresh chat", "summarize this so I can continue elsewhere", or otherwise signals a long conversation needs to carry forward into a new session without losing context.

  • A11y agent

    badass-courses/a11y-agent/.claude/skills/a11y-agent Skill

    no license0★ repo

    Accessibility expertise for building inclusive web interfaces. Covers semantic HTML, ARIA, keyboard interaction, screen reader support, color contrast, focus management, responsive zoom, motion safety, automated testing, and organizational culture. Use when building or reviewing any web UI to ensure it works for everyone.

  • Create subagent

    LoogacyStudio/skills/.agents/skills/create-subagent Skill

    0★ repo

    Use when an agent must decide whether a task should stay single-agent or be compiled into a reusable multi-agent topology with explicit subagent roles, context boundaries, orchestration contracts, and provider-specific implementation overlays.

  • My agent file conventions

    alexleekt/agents/skills/my-agent-file-conventions Skill

    no license0★ repo

    **ALWAYS use when:** creating, editing, or reviewing AGENT.md or claude.md files. **ALWAYS use when:** deciding what belongs in agent configuration vs project docs. **ALWAYS use when:** the user says "agent file", "agent config", "behavioral rules", "claude.md", "AGENT.md", "what goes in agent files", or "agent guidelines". **DO NOT use for:** tech stack recommendations (@skills/my-tech-stack) or VCS workflows. Defines the boundary: behavioral rules (agent files) vs documentation (README, ARCHITECTURE). Covers file locations, templates, common mistakes, and the "does this belong?" decision tree.

  • Sherma

    MadaraUchiha-314/sherma/skills/sherma Skill

    0★ repo

    Build LLM-powered agents with sherma — declarative YAML or programmatic Python, with multi-agent orchestration, skills, hooks, and A2A integration.

  • Gemini agent

    ninjaproger/skills/skills/gemini-agent Skill

    no license0★ repo

    Delegate complex, multi-step coding tasks to the Google Gemini CLI agent running in the background. Use when the user wants to offload autonomous coding work to gemini — including TCA feature development (new reducer/view/tests in a modular iOS app) and iOS Simulator automation (build, install, navigate, inspect). Gemini runs non-interactively via `gemini "prompt"`, handles tool calls and file edits autonomously, and returns a final summary. Use for tasks that benefit from background execution: "implement X feature using gemini", "have gemini add tests for Y", "use gemini to navigate the simulator and verify Z", "run gemini in the background to build this".