agentsclimarketplace

Context & AI engineering

25,991 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

  • Wundervault mcp

    wundervault/wundervault-mcp/io.github.wundervault/wundervault-mcp MCP server

    2★ repo

    Zero-knowledge MCP secrets vault for AI agents: secrets injected at runtime, never seen by the model

  • Action1

    wyre-technology/action1-mcp/io.github.wyre-technology/action1-mcp MCP server

    2★ repo

    Action1 MCP server — endpoint inventory, patch visibility, and policy automation via REST API.

  • Datto SaaS Protection

    wyre-technology/datto-saas-protection-mcp/io.github.wyre-technology/datto-saas-protection-mcp MCP server

    2★ repo

    MCP server for Datto SaaS Protection — M365/GWS backups, restores, seats.

  • QuickBooks Online

    wyre-technology/qbo-mcp/io.github.wyre-technology/qbo-mcp MCP server

    2★ repo

    MCP server for QuickBooks Online — accounts, customers, invoices, bills, and reports.

  • X402search mcp

    x402-index/x402search-mcp/io.github.x402-index/x402search-mcp MCP server

    no license2★ repo

    Search 13,000+ x402-enabled APIs from your AI agent. Powered by x402search.xyz

  • Helldivers2 mcp

    xerno42/helldivers2-mcp/io.github.xerno42/helldivers2-mcp MCP server

    2★ repo

    MCP server exposing live Helldivers 2 galactic war data.

  • Campus Copilot MCP

    xiaojiou176-open/campus-copilot/io.github.xiaojiou176-open/campus-copilot-mcp MCP server

    2★ repo

    Read-only MCP server for the local Campus Copilot BFF and imported workspace snapshots.

  • NoteStore Lab MCP

    xiaojiou176-open/apple-notes-forensics/io.github.xiaojiou176-open/notestorelab-mcp MCP server

    2★ repo

    Local copy-first Apple Notes recovery and review MCP server for NoteStore Lab case roots.

  • SourceHarbor MCP

    xiaojiou176-open/sourceharbor/io.github.xiaojiou176-open/sourceharbor-mcp MCP server

    2★ repo

    Source-first AI knowledge control tower MCP server for YouTube, Bilibili, RSSHub, and RSS.

  • Prompt Switchboard MCP Sidecar

    xiaojiou176-open/multi-ai-sidepanel/io.github.xiaojiou176-open/prompt-switchboard-mcp MCP server

    2★ repo

    Compare-first local MCP sidecar for browser AI workflows.

  • Xaip mcp server

    xkumakichi/xaip-protocol/io.github.xkumakichi/xaip-mcp-server MCP server

    2★ repo

    AI agents get on-chain identity, credentials, reputation, escrow, and persistent memory on XRPL.

  • tc39-mcp

    xyzzylabs/tc39-mcp/io.github.xyzzylabs/tc39-mcp MCP server

    2★ repo

    Independent MCP server for the TC39 specs (ECMA-262 + ECMA-402): clauses, search, diffs, history.

  • wasm-mcp

    xyzzylabs/wasm-mcp/io.github.xyzzylabs/wasm-mcp MCP server

    2★ repo

    Read-only MCP server for the WebAssembly spec: instructions, types, sections, search, proposals.

  • Siren

    yoavhizkiH/siren/io.github.yoavhizkiH/siren MCP server

    2★ repo

    MCP server for Siren diagramming — teach AI models to create beautiful diagrams with Siren syntax

  • Tablecloth mcp

    yourtablecloth/TableClothMcp/io.github.yourtablecloth/tablecloth-mcp MCP server

    2★ repo

    Open Korean e-Gov and finance sites in a clean, disposable Windows Sandbox with security programs.

  • Koncept mcp server

    yourtechtribe-labs/koncept-mcp/io.github.yourtechtribe-labs/koncept-mcp-server MCP server

    2★ repo

    Query a curated YAML semantic concept graph for codebase invariants and cross-cutting facts.

  • Opn mcp

    ysalitrynskyi/opn-mcp/io.github.ysalitrynskyi/opn-mcp MCP server

    2★ repo

    MCP server for opn.onl: shorten links, analytics, branded QR codes, and link management.

  • Thoth mcp

    perminder-klair/thoth-mcp/io.github.zeiq-co/thoth-mcp MCP server

    no license2★ repo

    MCP server for Thoth with multi-platform AI content generation

  • iFlow Search

    zhengyanglsun/iflow-search-js/io.github.zhengyanglsun/iflow-search MCP server

    2★ repo

    MCP server for iFlow Search web search, image search, and URL fetch tools.

  • Concord mcp

    zkwasm/concord-plugin/io.github.zkwasm/concord-mcp MCP server

    2★ repo

    Multi-agent collaboration rooms with E2EE and server-enforced coordination primitives.

  • BankBridge

    bankbridge-money/bankbridge-plugin/money.bankbridge/server MCP server

    2★ repo

    Read-only bank access for your AI agent. Connects Claude, ChatGPT, Cursor, Gemini, Codex.

  • Toolify

    sleetcrash/sleetcrash-skills/skills/toolify Skill

    1★ repo

    Mines the current session for reusable tool candidates (skill, agent, hook, slash command, automation, MCP) and triages each into build, defer, or drop. Invoke ONLY when the user explicitly types /toolify; do not auto-trigger this proactively.

  • Project kickoff

    magi12345/project-kickoff/project-kickoff Skill

    1★ repo

    Interview-style skill for starting a new project. It clarifies the goal and scope, then creates three tracking files (PROJECT.md, DECISIONS.md, PROGRESS.md) that act as persistent memory across all future Claude sessions. Use this skill EVERY time the user starts a new project of any kind — an app, a website, a plugin, a design, a prototype, a documentation or knowledge system, anything. Typical trigger phrases: I'm starting a new project, let's build something new, I want to make a website/app/plugin/tool, let's kick off, new project. At the end of the interview the skill doesn't just ask questions — it creates the three tracking files, writes the tracking convention into the project's CLAUDE.md, and from then on every conversation automatically records important decisions and completed units of work.

  • Claude md architect

    aimasteracc/harness-kit/skills/claude-md-architect Skill

    no license1★ repo

    Generate a constraint-first CLAUDE.md (agent operating-rules file) for the current repository by inspecting the project and interviewing the user about the rules that actually matter. Use when the user says "set up CLAUDE.md", "write agent rules for this repo", "claude-md architect", or onboards a new project to Claude Code.

  • Agent lint

    broneq/bdk/.claude/skills/agent-lint Skill

    1★ repo

    Lint a BDK agent file for best practices and portability. Use when reviewing, creating, or editing an agent (agents/*.md) to verify it meets BDK conventions. Trigger on phrases like "check this agent", "verify agent", "lint agent", or when about to commit a new agent.

  • Agent builder

    jafini/learn-claude-code/skills/agent-builder Skill

    1★ repo

    Design and build AI agents for any domain. Use when users: (1) ask to "create an agent", "build an assistant", or "design an AI system" (2) want to understand agent architecture, agentic patterns, or autonomous AI (3) need help with capabilities, subagents, planning, or skill mechanisms (4) ask about Claude Code, Cursor, or similar agent internals (5) want to build agents for business, research, creative, or operational tasks Keywords: agent, assistant, autonomous, workflow, tool use, multi-step, orchestration

    The same file is in 1 other repository

  • Emperor time

    arbi-elezi/emperor-time Skill

    1★

    Emperor Time — unified full-rigor coding discipline. Runs every task through a micro-waterfall SDLC (requirements → design → build → verify → deliver) with hard gates, scientific-method fact-checking (claim ledgers, prediction-before-test), mandatory self/hetero critique, "dowsing" to discover tasks and scan the machine for other coding agents (Claude Code, Kimi CLI, Codex, Copilot CLI, opencode, Ollama), consent-based orchestration of those agents, and Chain Jail capture of missing skills from the web. Use when the user says "emperor time", asks for maximum rigor, zero-hallucination delivery, multi-agent orchestration of local CLIs, or asks the agent to find its own work.

  • Summon

    secondorderai/summon-skill/skills/summon Skill

    1★ repo

    Generate bespoke, stack-aware Claude Code agent files for tech teams. Creates specialized AI agent personas (backend, frontend, fullstack, AI/ML, DevOps, QA, security, code reviewer, performance engineer, system architect, tech lead) that Claude Code autonomously picks up from `.claude/agents/`. Use this skill whenever the user says things like "create an agent", "I need a code reviewer agent", "generate a backend agent for this project", "set up agents for my team", "build me a QA specialist", "create a security auditor agent", "I want a tech lead agent", "set up an agent team", "create agents for this repo", or any request involving creating Claude Code agent definitions, AI specialist personas, or team compositions for agentic coding workflows. Also trigger when the user mentions "agency agents", "agent roster", "agent team", or wants to scaffold autonomous coding assistants into their project.

  • Mcp tactics

    nlink-jp/skills-series/mcp-tactics Skill

    no license1★ repo

    Choose the right nlink-jp MCP server for the situation, and call them in the right order. Use when investigating an IP address, domain, URL, MAC address / BSSID, or a pcap capture; when analysing a CSV/JSON/JSONL/Parquet file or writing throwaway Python for data; when producing narrated Japanese audio, a presentation video, or a locally generated image; or when a second opinion from another model would help. Also for 調査・トリアージ・不審IP・不審URL・不審メール・パケット解析・データ分析・ナレーション音声・解説動画・画像生成・セカンドオピニオン. Read this before reaching for any in-house MCP server, and especially before any lookup that could touch the party under investigation.

  • Live docs lookup

    dhegde11/live-docs-lookup Skill

    1★

    Use when building with any AI SDK or API — Anthropic, OpenAI, or Google. Fetches current documentation in real time before answering, so stale training data doesn't cause bugs. Trigger on: model selection, tool use, function calling, streaming, prompt caching, batch processing, SDK setup, or any AI API integration, planning, debugging, or code review.

  • Eval

    croftspan/gigo/.claude/skills/eval Skill

    1★ repo

    Test whether your assembled context actually improves AI output. Runs tasks bare vs assembled, uses comparative judging, reports the delta. Use gigo:eval when you want to prove your setup works or debug why it doesn't.

  • Delegation protocol

    jeroromano/claude-code-orchestration-template/.claude/skills/delegation-protocol Skill

    1★ repo

    Protocol for delegating implementation, investigation or review work to Codex or to local agents (fast-worker, deep-reasoner, diff-reviewer). Use it whenever work is about to be delegated, a review is needed before a merge, a Codex command fails or hits quota, or the user asks to implement a spec, review a branch, or split work between models - even if they never say "delegate".

  • Doc compact

    x0c/doc-skills/doc-compact Skill

    1★ repo

    文档整理与压缩。核心职责是在不丢行为信息的前提下压缩冗余文档——删复述、删死链、删历史沿革、收敛重复提醒;配套完成结构整理(纠正文档放置、重建根 AGENTS.md 索引、补 inline 指针、规范 CLAUDE.md 为单行 @*.md、判定并拆分过大项目的二级索引)和全局文档管理规范的检查/安装。当文档冗余膨胀、索引失效、AGENTS.md 膨胀、CLAUDE.md 混入杂质,或需要确认全局规范是否写清文档管理模式时使用。

  • Agent workflow

    Chrike/coding-agent-skills/skills/agent-workflow Skill

    no license1★ repo

    Lightweight Claude Code skills for daily AI-assisted development workflows.

  • Multi agent system architect v2

    ingcontartese-netizen/multi-agent-system-architect-v2/skills/multi-agent-system-architect-v2 Skill

    1★ repo

    Design, audit, build, verify, or port governed agentic systems for one host or optional teams. Triage no-system, workflow, agents; govern authority, state, tools, security, evals. Not for one-offs.

  • Coding llm council

    nikola-wd/vibe-agentic-coding-llm-council-skill/skills/coding-llm-council Skill

    no license1★ repo

    Convene an internal "council" of distinct senior-engineer personas to analyze a problem, stress-test an approach, and produce an implementation plan before writing code — grounding the decision in up-to-date online research (current library/API versions, vendor pricing and limits, regulatory rules) rather than stale training data. Use this whenever the user is analyzing a system, weighing architectural or design trade-offs, planning a non-trivial feature or refactor, evaluating a library/tool/vendor choice, debugging something whose root cause is unclear, or asks to "think through", "plan", "analyze", "research", "review the approach for", or "figure out how to build" something — even if they don't say the word "council". Trigger it especially for decisions that are expensive to reverse (data models, API contracts, infra, auth, migrations) or that depend on external facts that may have changed. Do NOT trigger for trivial one-liners, straightforward lookups, or when the user just wants code written for an already-decided approach.

  • Handoff resume

    marsmike/agentic-toolkit/plugins/handoff/skills/handoff-resume Skill

    1★ repo

    Resume from the latest handoff in this repo — load it and continue from the next step. Also surfaces any newer PreCompact auto-snapshot.

  • Dual build

    cjcsecurity/claude-code-dual-build/skills/dual-build Skill

    1★ repo

    Orchestrate a parallel multi-agent build with mandatory cross-review across Claude and Codex. Splits a multi-component coding task into ~equal Claude and Codex subtasks, runs them in parallel in isolated git worktrees, then has the OPPOSITE model review each diff before consolidation. Use for multi-file features, refactors with parallel slices, batch fixes, or audit-and-fix passes — bail to a single-agent build for trivial changes. The cross-review is the value: different model families catch different bug classes.

  • Analyst modes

    odere-pro/claude-wiki-pages-plugin/skills/analyst-modes Skill

    1★ repo

    The five operating modes (Query, Dashboard, Document Compile, Extract, Challenge) and the two write-gates for the claude-wiki-pages analyst. Documents each mode's procedure, output shape, and metric catalog, plus the dashboard/synthesis approval gates, so the analyst agent can run a mode without inlining the detail. Trigger when an agent or user asks "how does the analyst query/build a dashboard/compile/extract/challenge", "what are the analyst modes", or invokes /claude-wiki-pages:analyst-modes. Reference, not action.

  • Prompt coach

    aroyburman-codes/compound-product-management/skills/prompt-coach Skill

    1★ repo

    Use when the user's prompt is vague, overloaded, or missing context that would improve the output. Also use retroactively at session breakpoints to surface prompting patterns the user can improve. Coaches without being annoying.

  • Capturing session learnings

    pgoell/pgoell-claude-tools/plugins/agent-system-management/skills/capturing-session-learnings Skill

    1★ repo

    Capture session learnings into AGENTS.md or CLAUDE.md. Triggers on /revise-agents-md and /revise-claude-md.

  • Newproject

    cabaynes/charles-claude-skills/skills/newproject Skill

    1★ repo

    Use when bootstrapping a new project under a configured workspace directory ($WORKSPACE_DIR, default ~/projects/), or when an existing project folder needs the standard infrastructure (CLAUDE.md stub, memory dir, optional git init + GitHub repo, optional umbrella entry). Triggers only when the user references a project folder under the workspace — do NOT use for generic git init, npm/Xcode init, or single-file creation. Idempotent — never overwrites existing user content.

  • Fable prompt

    Amit-Tabibi/claude-code-skills/plugins/fable-prompt/skills/fable-prompt Skill

    1★ repo

    Use when the user needs a prompt written FOR Claude Fable 5 (or Mythos 5) — triggers on "I need a Fable 5 prompt", "prompt for Fable 5", "write me a Fable 5 prompt", "turn this into a Fable 5 prompt", "make a prompt for Fable". Takes a rough ask and returns a finished, Fable-5-optimized prompt to paste. NOT for prompting other Claude models or for general content writing.

  • Simple ai harness blueprint

    GuiomeB/simple-ai-harness-blueprint Skill

    1★

    Install, audit, or extend an AI-collaboration blueprint on a code repository — the AGENTS.md / CLAUDE.md / .agents/ structure that lets AI agents (Cursor, Claude Code, Codex Desktop/CLI, Windsurf) collaborate without drift. Three additive sizes (S/M/L) plus an opt-in L+ autonomy profile. Trigger when bootstrapping a new repo's AI scaffold, promoting an existing harness to the next size, or auditing one for sprawl.

  • Career kit orchestrator

    koralbonfil-maker/career-kit/skills/career-kit-orchestrator Skill

    1★ repo

    Your career journey guide — manages the full job search process from self-discovery to signed offer. Use this skill whenever someone wants to start their job search, go through the full career kit, get guidance on what to do next, or says anything like 'I want to start looking for a job', 'help me with my job search', 'what should I do first', 'guide me through the process', 'I'm starting to look for work', 'where do I begin', or 'what's my next step'. This is the starting point for anyone using this Career Kit — it guides them through each skill in the right order, tracks their progress, and makes sure nothing falls through the cracks.

  • Agent backup

    quietdaemon/starter-skills/skills/agent-backup Skill

    1★ repo

    Mirror your AI assistant's live config (skills, persona, settings, custom plugins) to a private git remote. Run on a schedule via cron — secrets and runtime state are excluded by an allowlist .gitignore. Use when asked for a manual backup, or when a backup cron fires.

  • Prompt coach

    goingli0324/prompt-coach Skill

    1★

    Turns a vague, underspecified request into a clearly structured Goal / Constraints / Obstacle breakdown, then produces a ready-to-use structured prompt. Use this whenever the user brings a fuzzy ask ("幫我想辦法...", "有什麼方法可以...", "我不知道怎麼處理..."), asks explicitly to help them "問對問題" or structure their request, or describes a real-world problem/decision they want AI help with but haven't yet specified what success looks like, what can't change, or what the actual blocker is. Trigger this proactively — even without the user naming the skill — any time answering directly would likely produce a generic, "technically correct but not actually usable" answer rather than something tailored to their real situation.

  • Cdfi peer benchmark

    Jaypatel1511/cdfi-superpowers/skills/cdfi-peer-benchmark Skill

    1★ repo

    Benchmark a bank CDFI against a peer group on FDIC call-report metrics (NIM, ROAA, ROAE, efficiency ratio, Tier 1 capital, loans-to-deposits, NPL ratio, loan-loss coverage). Use when the user says "benchmark this CDFI", wants a "peer comparison", or asks "how does this bank CDFI compare". Bank CDFIs only (FDIC-insured) — no credit unions, no unregulated loan funds. Backed by the audited PyPI package cdfi-benchmark; import name is `cdfibenchmark`.

  • Agent skill auditor

    fevziegeyurtsevenler/llm-security-skills/skills/agent-skill-auditor Skill

    1★ repo

    Statically audit any Agent Skill, MCP server config, or rules file (CLAUDE.md, .cursorrules, AGENTS.md) with the `uncloak` scanner BEFORE installing, enabling, or trusting it — catches invisible-Unicode instruction smuggling, MCP tool-description poisoning, credential/exfil paths, and lethal-trifecta posture. Use whenever you are about to add, copy, or fetch a third-party agent extension.

  • Design handoff

    humbleteam/design-handoff Skill

    1★

    Turns a mockup, screenshot, or HTML file into a dev-ready handoff spec - layout measurements, a token map, a component-state inventory, accessibility notes, and numbered open questions. Use for "write a handoff spec for this screen", "turn this mockup into dev docs", "document component states", or "what tokens does this screen use". Not for critiquing design quality or generating new UI - see design-review or html-mockup.

  • Prompt gatekeeper

    posixplus/prompt-gatekeeper/skills/prompt-gatekeeper Skill

    1★ repo

    Evaluate and enhance a prompt before executing it. Scores the request against a 6-pillar rubric (Role, Task, Context, Process, Guardrails, Output), classifies prompt depth to recommend a model tier, asks targeted clarifying questions for real gaps, then builds and runs a token-lean enhanced version of the prompt after the user confirms. Use this whenever the user includes [gatekeeper] or [Use Prompt Gatekeeper] in their message, asks to "improve", "enhance", "review", "score", or "gatekeep" a prompt, asks which model a task needs, or submits a broad, multi-step, open-ended request that lacks the constraints needed to get a high-quality result on the first try (e.g. "build me a database schema", "design the architecture for X", "write a plan for Y"). Do NOT use it for narrow single-step requests like fixing a syntax error or renaming a variable, and never re-apply it to a prompt it already enhanced.

  • New agent

    hcussi/claude-code-toolkit/.claude/skills/new-agent Skill

    no license1★ repo

    Scaffold a new agent in this claude-code-toolkit repo (create agents/<name>.md, the mirrored docs/agents/<name>.md, and add the README table row) following repo conventions. Use when adding a new agent to the toolkit.

  • Handoff pill

    Swarno1996/Handoff-pill Skill

    1★

    Generate a handoff pill, a single dense markdown file that carries a long chat's full context, decisions, solved problems, and the user's working directives into a fresh chat. Use whenever the user says pill, handoff pill, context pill, or ingestible pill; asks to package, transfer, or hand off this conversation to a new chat; says the chat is getting too long or slow and wants to continue elsewhere without losing context; or asks to distill the session's learnings into a portable file. Also use in reverse, whenever a file named like *_PILL.md or *HANDOFF*.md is uploaded or pasted at the start of a chat, by following this skill's ingestion protocol before doing anything else.

  • Graph engineering

    BhaveshKhaple/bhavesh-claude-skills/graph-engineering Skill

    1★ repo

    Design resilient multi-step LLM/agent workflows using the loops-and-graphs pattern. Covers all 5 layers: reflection, tool use, planning, multi-agent coordination, and critique loops — with anti-patterns and cost guidance.

  • Gdr prompt

    vintagedon/claude-skills-cookbook/skills/gdr-prompt Skill

    no license1★ repo

    Create effective Gemini Deep Research (GDR) prompts using Negative Space Bounding (NSB) methodology. Use when user needs to write a research prompt for GDR, wants to bound an agentic research task, or mentions NSB, "deep research prompt", or research that keeps wandering off-topic. Applies the three-layer constraint architecture (Anchors → Walls → Gates) to collapse latent space before research begins.

  • Claude code plugins

    JamesPrial/claude-plugins/.claude/skills/claude-code-plugins Skill

    1★ repo

    Contains all relevent documentation for Claude Code Plugins. ALWAY PROACTIVELY use when work on or with plugins.

  • Universal framework os

    TELLEBO/universal-framework-os Skill

    1★

    Universal operating system for every Claude interaction. Apply on every message regardless of topic — coding, writing, research, math, physics, scraping, planning, debugging, creative work, analysis, conversation, or any other task. Defines how Claude thinks, plans, executes, and validates ALL work using frameworks from 19 repos and 3 reference books. Triggers on everything: any prompt, question, request, topic, domain. Includes spec-driven development (GSD), agentic skills (Superpowers), agent harness (ECC), multi-agent swarm orchestration (Ruflo), web scraping (Scrapling, Crawl4AI, Firecrawl, Scrapy, Crawlee), intelligence pipelines (Crucix), autonomous research (AutoResearch), swarm prediction (MiroFish), GraphRAG knowledge graphs, cybersecurity (HexStrike), session memory (claude-mem), and math/physics reference knowledge. NOT optional — foundational layer for all work.

  • Wemush mycology assistant

    wemush/wemush-skills/wemush-mycology-assistant Skill

    1★ repo

    Guide mycology research workflows using WeMush's MCP server for specimen tracking, research projects, cultivation analytics, and strain catalog exploration. Use when users ask about mushroom cultivation, research projects, specimen management, or mycology data analysis.

  • Skill creator

    Modellix/modellix-skill/.agents/skills/skill-creator Skill

    1★ repo

    Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.

    The same file is in 66 other repositories

  • KanbanBoard

    robrohan/claude-skills/skills/KanbanBoard Skill

    1★ repo

    Manage an Obsidian Kanban board through a multi-agent workflow. Subcommands: init, plan, manage, work, verify, view.