agentsclimarketplace

Context & AI engineering

25,745 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

  • Improve agent

    stempeck/agentfactory/internal/cmd/install_skills/improve-agent Skill

    3★ repo

    Improve an agent's formula TOML based on post-execution learnings. Use after a formula run required manual intervention, produced incorrect artifacts, or left cleanup work for the operator. Classifies the failure type, scans for sibling vulnerabilities, selects the appropriate fix pattern, and surgically inserts corrective steps into the formula.

  • Atlassian mcp

    zacklecon/claude-skills/skills/atlassian-mcp Skill

    3★ repo

    Use when querying Jira issues, searching Confluence pages, creating tickets, updating documentation, or integrating Atlassian tools via MCP protocol.

  • Mcp developer

    zacklecon/claude-skills/skills/mcp-developer Skill

    3★ repo

    Use when building MCP servers or clients that connect AI systems with external tools and data sources. Invoke for MCP protocol compliance, TypeScript/Python SDKs, resource providers, tool functions.

  • Prompt engineer

    zacklecon/claude-skills/skills/prompt-engineer Skill

    3★ repo

    Use when designing prompts for LLMs, optimizing model performance, building evaluation frameworks, or implementing advanced prompting techniques like chain-of-thought, few-shot learning, or structured outputs.

  • Rag architect

    zacklecon/claude-skills/skills/rag-architect Skill

    3★ repo

    Use when building RAG systems, vector databases, or knowledge-grounded AI applications requiring semantic search, document retrieval, or context augmentation.

  • Subagent driven change

    jsuvic/agent-profile/.agents/skills/subagent-driven-change Skill

    3★ repo

    Use when a scoped implementation can be delegated to an implementation subagent and then independently reviewed for spec compliance before code quality.

  • Subagent driven change

    jsuvic/agent-profile/fixtures/logging-enforcement-enabled/expected/.agents/skills/subagent-driven-change Skill

    3★ repo

    Use when a scoped implementation can be delegated to an implementation subagent and then independently reviewed for spec compliance before code quality.

  • Subagent driven change

    jsuvic/agent-profile/fixtures/logging-enforcement-enabled/expected/.claude/skills/subagent-driven-change Skill

    3★ repo

    Use when a scoped implementation can be delegated to an implementation subagent and then independently reviewed for spec compliance before code quality.

  • Mcp fit check

    jsuvic/agent-profile/fixtures/mcp-recommendations-enabled/expected/.agents/skills/mcp-fit-check Skill

    3★ repo

    Use to assess whether current documentation through an MCP, docs, or search integration may help with a dependency that may be newer than the client's knowledge.

  • Mcp fit check

    jsuvic/agent-profile/fixtures/mcp-recommendations-enabled/expected/.claude/skills/mcp-fit-check Skill

    3★ repo

    Use to assess whether current documentation through an MCP, docs, or search integration may help with a dependency that may be newer than the client's knowledge.

  • Subagent driven change

    jsuvic/agent-profile/fixtures/subagent-driven-development/expected/.agents/skills/subagent-driven-change Skill

    3★ repo

    Use when a scoped implementation can be delegated to an implementation subagent and then independently reviewed for spec compliance before code quality.

  • Subagent driven change

    jsuvic/agent-profile/fixtures/subagent-driven-development/expected/.claude/skills/subagent-driven-change Skill

    3★ repo

    Use when a scoped implementation can be delegated to an implementation subagent and then independently reviewed for spec compliance before code quality.

  • Mcp fit check

    jsuvic/agent-profile/fixtures/tabnine-workflow-skills/expected/.agents/skills/mcp-fit-check Skill

    3★ repo

    Use to assess whether current documentation through an MCP, docs, or search integration may help with a dependency that may be newer than the client's knowledge.

  • Delegation scorer

    sohaibt/founder-mode/skills/delegation-scorer Skill

    3★ repo

    Score whether a delegation decision is healthy (amplifying what you love) or harmful (abdicating what you don't understand). Based on Chesky's principles on what to keep vs. what to let go. Use when deciding whether to delegate something or stay involved.

  • Run benchmark

    tmuskal/arc-agi-benchmarker/plugins/arc-agi-benchmarker/skills/run-benchmark Skill

    no license3★ repo

    Execute benchmark runs against ARC-AGI games - plays games with Claude Code as the agent and records scores

  • Run benchmark

    tmuskal/arc-agi-benchmarker/plugins/longmemeval-benchmarker/skills/run-benchmark Skill

    no license3★ repo

    Execute a LongMemEval benchmark run - drives per-item hypothesis generation and LLM-as-judge scoring with checkpointing, resumable, capped at 500 items by default

  • Source command job eval

    cyn-zhang/job-radar/.agents/skills/source-command-job-eval Skill

    3★ repo

    Evaluate a JD — requirements, ATS keywords, fit verdict

  • Review loop

    NYTC69/review-loop/skills/review-loop Skill

    3★ repo

    Automates a Plan-Execute-Review workflow with built-in iteration loops. An Orchestrator coordinates an Executor sub-agent and a configurable Reviewer (external AI CLI like Codex, or Claude sub-agent) to drive work items from description to delivery — with every review finding visible to the user. Trigger: "run review-loop on", "start review loop", "let the agents handle", or any task the user wants driven by a plan→review→implement→CR cycle.

  • New mcp tool

    justnau1020/claude-os/skills/development/new-mcp-tool Skill

    3★ repo

    Add a new MCP tool to a FastMCP server. Use when adding a new tool for LLM agents or extending the MCP interface with new functionality.

  • Memory checkpoint

    u9401066/template-is-all-you-need/.claude/skills/memory-checkpoint Skill

    3★ repo

    Externalize detailed memory to Memory Bank before conversation summarization to prevent context loss. Triggers: CP, checkpoint, save, 存檔, 記一下, 保存, sync memory, dump, 先記著, 備份, backup, snapshot, 快照, 記錄, 等等, 先停, pause, 離開前, before leaving, 收工.

  • Memory updater

    u9401066/template-is-all-you-need/.claude/skills/memory-updater Skill

    3★ repo

    Update and maintain Memory Bank files (activeContext, progress, decisionLog). Triggers: MB, memory, 記憶, 進度, 更新記憶, update memory, 記錄進度, 更新上下文, sync, 同步, 記下來, note, 筆記, context, 脈絡, 追蹤, track, 狀態, status.

  • Index

    thoreinstein/obsidian-rag/skills/index Skill

    3★ repo

    Use when the user wants to index their vault for semantic search, rebuild the RAG index, or says "index", "reindex", or "rebuild index".

  • N8n syntax ai nodes

    Impertio-Studio/n8n-Claude-Skill-Package/skills/source/n8n-syntax/n8n-syntax-ai-nodes Skill

    3★ repo

    Use when building AI or LLM workflows in n8n v1.x (v1.19.4+). Prevents incorrect sub-node wiring by mismatching NodeConnectionTypes. Covers agent nodes (6 types), chain nodes, tool nodes, memory backends (8 types), vector stores (11 types), output parsers, text splitters, retrievers, AI sub-node connections (12 NodeConnectionTypes), langchain integration, RAG patterns, and human-in-the-loop. Keywords: n8n, AI nodes, LLM, langchain, RAG, vector store, agents, ChatGPT in n8n, AI workflow, LLM chain, RAG pipeline, vector search, AI agent..

  • Llm inference benchmark

    soulmachine/skills/llm-inference-benchmark Skill

    3★ repo

    Benchmark OpenAI-compatible LLM inference servers (vLLM, SGLang, or anything serving /v1/completions; local, cross-host via TARGET_HOST=<peer LAN IP>, or behind a TLS+Bearer proxy via --base-url + OPENAI_API_KEY) with sglang.bench_serving run as a standalone dockerized client — one engine-agnostic script (scripts/bench_sweep.sh), default sweep c1→c128 uncapped, sustained-load grid (uniform prompts-per-concurrency), self-describing PROVENANCE log headers — collecting TTFT, TPOT, ITL, and input/output/total token throughput. From those metrics, derive insights: (1) the min and max knee of the saturation curve — min = highest concurrency where TTFT/TPOT is still flat (latency-optimal cap), max = where output throughput goes flat (throughput ceiling) — to pick a production --max-num-seqs / --max-running-requests; (2) whether the box is prefill- or decode-bound (the min-vs-max-knee gap, IN/TTFT vs 1/TPOT, or a prefill-heavy/decode-heavy shape sweep); (3) whether a throughput plateau is a real compute/comm knee or a false one (a binding MAX_SEQS cap or KV exhaustion) — via the flat-throughput+flat-TPOT+rising-TTFT signature, the running-vs-offered gauges (num_running/num_waiting), and a raise-the-cap-and-re-measure check. Includes verified Kimi-K2.6 baselines on 8× RTX PRO 6000 Blackwell (INT4 SGLang/vLLM, fp8-vs-bf16 KV, NVFP4 marlin-vs-b12x, Caddy proxy overhead). Use when asked to benchmark or compare LLM servers (tokens/sec, TTFT, TPOT, ITL) across engines, quantizations, KV-cache dtypes or MoE backends, find the saturation knee / pick a concurrency cap, determine prefill- vs decode-boundedness, run a cross-host A/B, quantify reverse-proxy overhead, interpret why high-concurrency numbers look inflated (short-burst trap) or why two sweeps aren't comparable (MAX_SEQS / grid mismatch), or debug a sweep that prints SWEEP_DONE with no metrics.

  • Ssh claude auth

    soulmachine/skills/ssh-claude-auth Skill

    3★ repo

    Fix Claude Code appearing logged-out over SSH on a headless macOS machine — its credentials sit in the login keychain, which stays locked in SSH/headless sessions. Offers two fixes and asks the user to choose: a keychain-free long-lived OAuth token (recommended), or auto-unlocking the login keychain with a stored password. Use when Claude Code shows unauthenticated over SSH on a Mac, when setting up a Mac mini or headless Mac for remote/CI Claude Code use, or when `security show-keychain-info` reports the login keychain locked.

  • Goal workflow

    adamlinscott/claude-skills/skills/goal-workflow Skill

    3★ repo

    Run a settled implementation goal as a bounded autonomous build loop — lock the goal from the conversation and any docs written this session, front-load every decision, map the terrain, write a completion-invariant contract before any code, then loop (building with explicit Agent-tool fan-out, committing at intervals, verifying at milestones with fresh-eyes against the contract) until the invariants hold, and close out. Gated on a `--confirm` flag asserting the user has manually set ultracode effort and auto-accept mode (a skill can set neither); without the flag it stops, gives the setup steps, and offers two paths — a managed `--confirm` run, or a copy-pasteable `/goal` command that hands the work to native goal + workflow orchestration. An optional `--commit` flag (default off) enables commit-at-intervals and push-at-milestones in both the managed run and the generated command; without it, nothing is committed. Use when a plan is settled and you want Claude to implement it end-to-end on its own — typically after a planning skill — or whenever the user says goal-workflow.

  • Memory audit

    adamlinscott/claude-skills/skills/memory-audit Skill

    3★ repo

    Audit Claude's per-user memory for this project and produce a report. Asks at the start whether the user wants a non-technical summary or a full technical audit. Never edits memory — produces a report only. Usage: /memory-audit

  • Cost tier routing

    The-Artificer-of-Ciphers-LLC/skills-from-the-artificer/cost-tier-routing Skill

    3★ repo

    Use BEFORE doing direct file search, bulk reads, data import/export, or routine coding edits in the main conversation. Routes work to the cheapest model that can do it correctly — haiku for search/IO, sonnet for coding, opus for orchestration/architecture. Triggers when about to call Read on >2 files, Grep/Glob across the repo, batch CSV/JSON transforms, or any "where is X / list all Y / count Z" question. Also triggers when the orchestrator (you, on opus) is about to write straightforward code that a sonnet subagent could handle.

  • Trust but verify

    The-Artificer-of-Ciphers-LLC/skills-from-the-artificer/trust-but-verify Skill

    3★ repo

    Use when you receive findings, a report, or any factual claims back from a subagent/Task before you act on them — re-validate every claim against a primary source (the code itself, the memory directory, CONTEXT.md, docs/ and docs/adr/, context7 library docs, package/API references, or a language spec) instead of trusting the agent's summary. Trigger automatically after any Task/subagent returns a result you're about to rely on, and whenever the user says "trust but verify", "verify these claims", "double-check what the agent found", "did it actually confirm that", or "where's the source for X". A subagent's report is a lead, not a fact. Nothing is verified until a source was opened and quoted; guessing, inferring, and "that looks right" are not verification.

  • Tracegen matrix

    rashidee/co2-skills/skills/tracegen-matrix Skill

    3★ repo

    Requirement-to-code traceability generator — links every PRD requirement ID (User Stories, Non-Functional Requirements, Constraints) and bug code from a module's SPEC.md to the source code that implements it, and writes a per-application context/TRACEABILITY.md matrix. Resolves links primarily from the in-source CO2 traceability comments written by conductor-feature-develop and conductor-defect (e.g. `Implements: USHM00003`, `Bug fixes: [BUG-024]`), with a name-based source-scan fallback, and OPTIONAL precision enrichment from the codebase-memory knowledge graph when that MCP server is installed. Works fully WITHOUT codebase-memory. Takes an application name (mandatory), with optional version and module filters. Use when the user asks to "generate traceability", "update the traceability matrix", "link requirements to code", "regenerate TRACEABILITY.md", or when invoked as a sub-step by the feature/defect conductors after code or bug-fix changes.

  • Product pulse

    shwetank/bettersense/plugin/skills/product-pulse Skill

    no license3★ repo

    Use to produce a single-page dated pulse report on product health — usage metrics, system performance, AI-feature signals (hallucination, refusal, eval drift), counter-metrics, and follow-up investigation. Trigger phrases include "run a pulse on [area]", "give me a product health snapshot", "weekly product pulse", "what does the data say this week". Reads strategy.md to know what to watch and which counter-metrics matter. Saves to `~/bettersense-work-reflections/pulses/<area-slug>/pulse-YYYY-MM-DD.md` so the folder accumulates into product memory over time. Designed for daily/weekly Desktop scheduled tasks (not cloud routines — those cannot read or write local files). Privacy-first — defaults to aggregated/redacted data; surfaces provenance.

  • Docker orchestrator

    FutureJJ/claude-skills/skills/docker-orchestrator Skill

    3★ repo

    Docker containerization including multi-stage builds, Docker Compose, security best practices, image optimization, and orchestration patterns. Trigger when users create Dockerfiles, docker-compose.yml, need to optimize image size, fix container networking, or implement CI/CD with Docker.

    The same file is in 1 other repository

  • Food chain ideation

    CodedRichy/food-chain-ideation/skills/food-chain-ideation Skill

    3★ repo

    10-minute pre-ship stress-test. Run this before you ship, not instead of shipping. Dynamically selects animal agents from a behavioral DNA library matched to the specific problem, each attacking under strict role-lock with zero shared context. Weakest eliminated each round, survivor absorbs the insight and evolves, idea patches itself until one apex predator stands. Quick mode (5 animals, 1 round) for fast verdicts. Full mode for high-stakes decisions. Works in Claude.ai, Claude Code, Cursor, Windsurf, Copilot, and any agent with no dependencies.

  • Food chain monitor

    CodedRichy/food-chain-ideation/skills/food-chain-monitor Skill

    3★ repo

    Re-stress-tests an idea that was previously battle-tested with food chain. Takes a Battle Log from a prior session plus a description of what changed, then runs a focused battle targeting only the NEW assumptions introduced by the change. Does not re-test what already survived. Works in Claude.ai, Claude Code, Cursor, Windsurf, Copilot with no dependencies.

  • Food chain pitch

    CodedRichy/food-chain-ideation/skills/food-chain-pitch Skill

    3★ repo

    Investor pitch stress-tester. Spawns adversarial agents from a pitch-specific behavioral DNA library. Each attacks the pitch narrative, financial model, or due diligence surface under strict role-lock. Elimination rounds harden the pitch. Works in Claude.ai, Claude Code, Cursor, Windsurf, Copilot with no dependencies.

  • Huashu prompt save en

    Biraj2004/huashu-skills-english/huashu-prompt-save-en Skill

    3★ repo

    Automatically identifies the prompt type and saves it to the appropriate category (Technical / Content / Teaching / Product / General). Use when the user mentions "save prompt", "record prompt", or "organise prompts".

  • Prompt creator

    skizha/agent-skills/skills/prompt-creator Skill

    no license3★ repo

    Expert prompt engineering for AI models (especially Claude). Use when the user wants to: create a new prompt from scratch, improve or debug an existing prompt, design a system prompt for a chatbot or assistant, write few-shot examples, structure chain-of-thought reasoning, generate structured/JSON output, build RAG or agentic prompts, or evaluate why a prompt is underperforming and how to fix it. Triggers on requests like "write a prompt for...", "create a system prompt", "improve this prompt", "help me prompt engineer", "why isn't my prompt working", "make a prompt that outputs JSON".

  • Memory checkpoint

    Daisybastioned440/lite-research-agents/skills/memory-checkpoint Skill

    3★ repo

    Defines how to save and load agent state as checkpoint files.

    The same file is in 1 other repository

  • Memory checkpoint

    frankaging/lite-research-agents/skills/memory-checkpoint Skill

    3★ repo

    Defines how to save and load agent state as checkpoint files.

    The same file is in 1 other repository

  • Tech brief orchestrator

    harryl6798/vc-diligence-skill/skills/tech-brief-orchestrator Skill

    no license3★ repo

    High-level technical "Editor-in-Chief" that coordinates specialized sub-agents to produce exhaustive, research-grade technical briefs.

  • Handoff

    TitusKirch/skills/skills/work/handoff Skill

    3★ repo

    Hands in-progress work from one agent or session to another through a structured handoff document committed to the repo under `.agents/handoffs/`, and resumes an existing handoff by its number. Captures goal, context gathered, progress, next steps and open questions in a document that assumes none of the writer's session, so the work can continue in a fresh session or on another machine. The resuming agent deletes the handoff once the work is finished — not when it reads it. Use when the user wants to hand off, park or pause work for another agent or session, to write or read a handoff, to continue handed-off work, or says things like "hand this off", "write a handoff", "continue handoff 0003", "resume the handoff", "übergib das", "fahre das fort".

  • Ask claude

    mrzhangguoguo/oh-my-workbuddy/skills/ask-claude Skill

    3★ repo

    Deprecated compatibility shim for Claude advisor requests. Use `ask` with the claude backend instead.

  • Doctor

    mrzhangguoguo/oh-my-workbuddy/skills/doctor Skill

    3★ repo

    Diagnose and fix oh-my-workbuddy skill installation issues — missing or duplicate skill roots, stale package versions, broken AGENTS.md/memory. Trigger "doctor", "fix skill install", "why is my skill missing", "diagnose skills".

  • Swarm

    mrzhangguoguo/oh-my-workbuddy/skills/swarm Skill

    3★ repo

    Deprecated compatibility shim for team execution. Use `team` (Agent-tool coordination) instead.

  • Worker

    mrzhangguoguo/oh-my-workbuddy/skills/worker Skill

    3★ repo

    Protocol for a unit of work assigned to a WorkBuddy subagent (Agent tool) — ACK, claim-safe lifecycle, evidence-backed completion, and clean blocking escalation. Triggers: "worker protocol", "subagent task lifecycle", "claim task", "team worker".

  • Agent anatomy

    bowtiefunnel/bowtie-funnel-Labs/skills/agent-anatomy Skill

    3★ repo

    Use when scaffolding a new AI agent's project folder, or when an existing agent's files (prompt, tools, procedures, scheduled jobs, integrations) are scattered in one directory with no structure. Organizes any agent folder into a filesystem-first layout.

  • Pick agent stack

    bowtiefunnel/bowtie-funnel-Labs/skills/pick-agent-stack Skill

    3★ repo

    Use when choosing the infrastructure that runs an AI agent — "what should run this agent", "which execution engine / LLM router / eval / observability tool", "set up the agent stack", or after pick-gtm-agent-pattern settles the architecture and the runtime components aren't chosen yet. Selects per infrastructure layer from the Bowtie Funnel Agent Infrastructure stack (the Headless GTM OS) instead of guessing from training data.

  • Pick gtm agent pattern

    bowtiefunnel/bowtie-funnel-Labs/skills/pick-gtm-agent-pattern Skill

    3★ repo

    Use when planning a new GTM AI agent, turning an existing GTM workflow into an agent, or auditing a GTM agent for over-engineering — before an architecture is settled or a spec is written. Anchors the choice against the 46-agent Bowtie Funnel Model Switchboard fleet. Triggers on "plan this agent", "what pattern should this GTM agent use", "make this workflow an agent", "should this be an agent", "rebuild X agent", or "is this over-engineered".

  • Run an agent swarm

    protosphinx/sphinxstack/skills/run-an-agent-swarm Skill

    3★ repo

    Run several coding agents in parallel on one repository from a shared work queue, with per-agent isolation, serialized builds, disk limits, and a supervision loop that reclaims stalled work. Use when someone asks to run a coding swarm, work through open beads overnight, keep several agents moving through a backlog, or stop parallel agents from colliding in one checkout or filling the disk.

  • Bmad agent architect

    mannnrachman/pi-bmad/skills/bmad-agent-architect Skill

    no license3★ repo

    System architect and technical design leader. Use when the user asks to talk to Winston or requests the architect.

  • Bmad agent dev

    mannnrachman/pi-bmad/skills/bmad-agent-dev Skill

    no license3★ repo

    Senior software engineer for story execution and code implementation. Use when the user asks to talk to Amelia or requests the developer agent.

  • Bmad agent pm

    mannnrachman/pi-bmad/skills/bmad-agent-pm Skill

    no license3★ repo

    Product manager for PRD creation and requirements discovery. Use when the user asks to talk to John or requests the product manager.

  • Bmad agent tech writer

    mannnrachman/pi-bmad/skills/bmad-agent-tech-writer Skill

    no license3★ repo

    Technical documentation specialist and knowledge curator. Use when the user asks to talk to Paige or requests the tech writer.

  • Bmad agent ux designer

    mannnrachman/pi-bmad/skills/bmad-agent-ux-designer Skill

    no license3★ repo

    UX designer and UI specialist. Use when the user asks to talk to Sally or requests the UX designer.

  • Local file memory

    wchen02/cursor-agent-learning/.cursor/skills/local-file-memory Skill

    no license3★ repo

    Use a Markdown file in the repo as persisted agent memory across chats. Use when the user wants session-spanning context, a decision log, or to resume work without re-pasting history.

  • Mcp docs and implement

    wchen02/cursor-agent-learning/.cursor/skills/mcp-docs-and-implement Skill

    no license3★ repo

    Implement using up-to-date docs or APIs via MCP. Use when the user wants to build something with current library/API docs (e.g. Context7, Stripe) and MCP is available.

  • Setup

    skrrt-sh/skills/skills/ship/setup Skill

    3★ repo

    Configure this repository for the Skrrt ship workflows and select a branching strategy. Run once before commit, pr, and release.

  • Prompt structurer

    b-mendoza/agent-skills/skills/prompt-structurer Skill

    no license3★ repo

    A collection of reusable skills for AI-assisted development workflows

  • Subagent driven development

    b-mendoza/agent-skills/.agents/skills/subagent-driven-development Skill

    no license3★ repo

    Use when executing implementation plans with independent tasks in the current session

    The same file is in 6 other repositories

  • Workflow skill architect

    b-mendoza/agent-skills/skills/workflow-skill-architect Skill

    no license3★ repo

    A collection of reusable skills for AI-assisted development workflows