agentsclimarketplace

Context & AI engineering

25,864 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

  • Raise room orchestrator

    satishTheLegend/raise-room-orchestrator Skill

    0★

    Runs a startup fundraise end-to-end as one governed workspace — not just a pitch deck. Assesses raise-readiness and round shape; routes to existing deck skills for the narrative (never rebuilds them); builds and sanity-checks the financial model (burn, runway, cohorts, unit economics); ASSEMBLES a categorized data room and AUDITS a real folder for gaps before investors; runs the investor pipeline as a lightweight CRM with fit scoring and stall detection; prepares an adversarial diligence Q&A war room with every answer grounded in a real data-room artifact (flagging any unsupported claim); compares term sheets and models dilution from the cap table; and stands up a recurring investor-update engine. Keeps deck = model = data room consistent, and refuses to call a data room "investor-ready" while required corporate or financial artifacts are missing. Use whenever the user mentions fundraising, raising a round, pre-seed/seed/ Series A/B, pitching investors, a data room, due diligence, an investor pipeline, a term sheet, dilution, the cap table, runway, an investor update, or "who should I raise from" — even if they only ask for a pitch deck (route it through the deck phase) or never say "plan a raise". Trigger on one-liners like "we're raising a seed round" or "get diligence-ready".

  • Ai rmf governor

    satishTheLegend/ai-rmf-governor Skill

    0★

    Runs NIST AI RMF (Govern, Map, Measure, Manage) for a solo or small-team builder shipping an LLM/ML feature, turning "it seemed fine in the demo" into a living AI-RMF profile plus a runnable offline eval gate: named failure modes, chosen metrics with acceptable limits AND course-correction triggers, an explicit list of risks deliberately NOT measured and why, and signed-off residual risk. Use whenever the user is building, shipping, reviewing, or changing anything that calls an LLM or ML model — a chatbot, RAG pipeline, agent, classifier, summarizer, extraction step, prompt, or judge — EVEN IF they don't explicitly ask for governance, evals, or risk management. Use proactively when the user says they will "ship", "deploy", "it works", "good enough", or shows a model/prompt with no evals; when they SWAP a model, edit a prompt, or add a use case (re-open Map/Measure); when they mention hallucination, prompt injection, jailbreak, bias, PII/data leakage, over-reliance, cost blowup, or quality regression; or when they ask "is this safe to ship?". Defers generic risk registers to risk-register-csf and generic threat modeling to threat-model-studio, owning the AI-specific harms and eval rigor those do not. Iterative, not one-shot.

  • Whatsteam ai agent

    whatsteamhq/whatsapp-claude-skills/skills/whatsteam-ai-agent Skill

    0★ repo

    Build a production-grade AI agent that answers your customers on WhatsApp with WhatsTeam — an LLM (OpenAI/Claude) replies to inbound messages inside the 24-hour window, with a hard human handoff for sensitive or low-confidence cases, idempotent webhook handling, opt-out, and a kill switch. Use when the user wants an "AI chatbot / assistant / auto-responder" that handles FAQs, qualifies leads, and covers after-hours on WhatsApp — and wants it safe enough to put in front of real customers, not a naive autonomous bot.

  • System upgrade

    kazani-351/system-upgrade Skill

    0★

    Run high-leverage system-upgrade meta-prompts on the user's own agent setup and life/work system, not on application code. Use whenever the user wants to audit, optimize, tune, trim, or upgrade their AI harness or agent config (CLAUDE.md, AGENTS.md, hooks, skills, memory, system prompt); reconcile configs that drifted across multiple agents (Claude Code, Codex, etc.); harden against prompt injection or map their deployed attack surface; get strategic clarity on scattered projects or a career; stress-test a big bet before committing months to it; or find what would break if they vanished for a month. Triggers on phrases like "upgrade my system", "tune/audit my harness", "clean up my CLAUDE.md", "run the Fable prompts", or "optimize how you work for me". Goes one prompt at a time, applies reversible changes with diffs, confirms irreversible ones. Not for reviewing application code, a single config-line fix, or creating new skills. Works on any capable agent; benefits from maximum reasoning effort.

  • Orchestra memory

    luigipascal/skill-forge/.cursor/skills/orchestra-memory Skill

    no license0★ repo

    Queries Orchestra shared memory at localhost:4317 before coding when context may exist in prior chats, synced repos, or ingested threads. Use when the user mentions Orchestra memory, Untitled threads, recall, or cross-project context.

  • Codebase map

    peterphoenix/deep-reasoning/skills/codebase-map Skill

    0★ repo

    Generate and maintain persistent project knowledge (an agent memory file such as AGENTS.md or CLAUDE.md, plus ARCHITECTURE.md) so the agent starts every session already knowing the codebase. Use when: starting work in a repository that has no memory file or a stale one; the user says 'onboard', 'index this repo', 'remember this project', 'set up project memory', or 'update project context'; or after completing significant work that changed the architecture, conventions, or gotchas worth persisting. Also use proactively — if a session begins in a repo without a memory file, offer to generate one.

  • Gpt engineer spark

    SYMBaiEX/gpt-orchestration/skills/gpt-engineer-spark Skill

    0★ repo

    Lead end-to-end engineering work with a capable main agent and a parallel fleet of model-pinned GPT-5.3-Codex-Spark subagents. Use when the user asks for a Spark fleet, ultra-fast parallel coding agents, rapid repository exploration, many bounded implementation shards, or low-latency independent verification while retaining architecture, integration, and final acceptance with the main agent.

    The same file is in 1 other repository

  • J lens observability

    Gonzih/j-lens-observability-skill/j-lens-observability Skill

    0★ repo

    J-Lens and model reasoning observability for AI harnesses. Use when asked about Anthropic J-Lens/J-space/Jacobian lens research, hidden model thinking, prompt/response thought traces, chain-of-thought observability, reasoning audits, OpenClaw/Claude/Codex harness instrumentation, local session logs, or activation-level analysis on open-weight models.

  • Agent retrospective

    marcozhou26/agent-retrospective/agent-retrospective Skill

    0★ repo

    Diagnose and improve a Codex, Agent, Skill, or workflow after failure, repeated correction, false completion, local symptom fixes, rule stacking, or before promoting any learning into AGENTS.md, SKILL.md, scripts, dependencies, memory, or eval gates. Use when the user asks to review why a task went wrong, why a skill/tool/process still feels broken after one successful run, or how to improve the whole working system instead of adding another isolated rule.

  • Governed engineering runtime

    Abeer-ahmad-123/governed-engineering-runtime/skills/governed-engineering-runtime Skill

    0★ repo

    Operate a governed multi-agent software engineering runtime through a bootstrapped local ai CLI and the authenticated Codex CLI. Use when a repository change should pass through specification, architecture, specialist delegation, human approvals, implementation, independent review, security review, testing, persistent memory, and Git checkpoints rather than being edited immediately.

  • Agent browser webauthn

    kalepail/skills/skills/agent-browser-webauthn Skill

    0★ repo

    Use agent-browser with Chrome DevTools Protocol virtual WebAuthn authenticators to test passkey, WebAuthn, Stellar smart account, wallet creation, wallet restore, and browser signing flows. Trigger when a task mentions passkeys, WebAuthn, virtual authenticators, Stellar Smart Account Kit browser tests, or agent-browser passkey automation. Do not use for browser flows without a passkey or WebAuthn ceremony—password logins and extension-wallet connects route to agent-browser alone.

  • Headless cli

    oldantique/headless-cli-playbook/skills/headless-cli Skill

    0★ repo

    Call another local AI CLI headlessly from a Bash call — codex, opencode, agy, or kimi — to delegate a subtask, get a second opinion from a different model, fan out work in parallel, or run a one-shot prompt and capture the result. Use when the user asks to use/ask codex, opencode, agy, kimi, gemini, gpt, deepseek, or k3; wants another model's take or a cross-model check; when delegating coding/analysis to a non-Claude model is useful; or when the user wants an image generated (logo, banner, cover, illustration) — codex can do that headlessly. Contains the exact verified flags plus the gotchas (stdin/sandbox/parallelism) that otherwise cause hung or empty calls.

  • Senior prompt engineer

    Arushmukundan/senior-prompt-engineer/senior-prompt-engineer Skill

    no license0★ repo

    Transforms any vague, generic, or short user request into a comprehensive, senior-level prompt/spec BEFORE doing the work, then executes against that spec. Use this whenever the user asks you to build an app, write or refactor code, create a website or landing page, design UI/UX, make a presentation or slides, write content, do research, or analyze data - and especially any time a request feels underspecified. Also use it when the user says a previous result was wrong, not what they expected, or asks you to "try again" - so the skill can learn from the miss and self-correct.

  • Coding agent prompt optimizer

    ilovehugetits/coding-agent-prompt-optimizer/skills/coding-agent-prompt-optimizer Skill

    0★ repo

    Transforms rough, vague, conversational, or incomplete software-development requests into precise, execution-ready prompts for autonomous coding agents. Use when the user asks to optimize, rewrite, structure, improve, translate, or convert a coding task, bug report, feature note, or requirement into a prompt for Claude Code, Codex, Cursor, GitHub Copilot, Cline, Windsurf, or another repository-aware coding agent — including Turkish, English, or mixed-language input. Do not use for ordinary programming help such as explaining concepts, debugging code, or reviewing an implementation.

  • Skill miner

    Sagart-cactus/skill-miner/skills/skill-miner Skill

    0★ repo

    Mine past coding-agent session history (Claude Code, Codex CLI, opencode, pi, …) to discover and build the reusable skills a person or team should create. Use whenever the user wants to extract skills, conventions, or workflows from their session transcripts — "what skills should we create", "go through my past sessions", "turn my corrections into rules", "build a skill catalog" — or wants to run this across a whole team by collecting sanitized findings from each member's local history. Works for individuals and organizations alike.

  • Incident to eval

    hamidettefagh/incident-to-eval Skill

    0★

    Turns a production AI agent incident into a golden eval case, a portable regression test that makes the failure unrepeatable. Use when an agent did something wrong in production or testing, when someone has a postmortem, trace, bug report, or a paragraph describing an agent failure, when someone asks how to stop an agent failure from happening again, or when converting incidents, escalations, or wrong answers into evals, test cases, or a regression suite. Reconstructs the incident, classifies the failure against an agent-specific taxonomy, writes a vendor-neutral eval case, and validates it with a deterministic script.

  • Sync claude context

    TobiasWestholm/sync-claude-context Skill

    0★

    Sync all Claude Code project context at the start of a new session or after significant changes. Use this whenever: starting fresh on a project after a gap, asked to "sync", "catch up", "update project docs", "bring yourself up to speed", or when you suspect CLAUDE.md, skills, or memory might be stale. Proactively suggest it when the user says they've been away, a big feature landed, or the stack changed. Works for any language and project structure. Covers CLAUDE.md accuracy, project skills, memory consolidation, and settings review.

  • Perfect prompt

    takeshijuan/perfect-prompt/skills/perfect-prompt Skill

    0★ repo

    Use this skill whenever the user wants a terse request, slash command, vague task, issue or PR instruction, feature idea, UI/dashboard request, debugging task, QA request, release task, research request, planning prompt, or existing prompt rewritten into a stronger ready-to-paste agent prompt. Trigger even when the user only writes shorthand like "review PR#123", "implement login system", "add a dashboard view", "address issue #123", "debug staging auth", or "/perfect-prompt: ...". This skill generates the prompt only; it does not execute the task. Before composing, it gathers context: it reads the current conversation, reads configured user memory only when a memory system is detected, and resolves external references (issues, PRs, tickets, URLs, file paths) with available read-only tools so the generated prompt targets the real problem.

  • Mailkite

    mailkite/agent-skills/mailkite Skill

    0★ repo

    Set up and operate MailKite email end-to-end — create an account, add a sending/receiving domain, set its DNS at the user's provider (Cloudflare, GoDaddy, Namecheap, Route 53, …), register a webhook, design templates, send mail, and confirm inbound delivery. Use whenever the user wants to send email, receive email as a webhook, give an AI agent its own inbox, wire up a domain for email, or debug MailKite delivery. Works via the MCP server, the `mailkite` CLI, any language SDK, or raw REST.

  • Orchestrate

    gabros20/orchestrate-skill/skills/orchestrate Skill

    0★ repo

    Coordinate multi-agent work by selecting and running a bounded orchestration strategy. Use for implementation plans or multi-step tasks needing staged reviews, parallel worktrees, hierarchical fleets, agent teams, workflows, goal loops, advisor/executor splits, adversarial planning, or external coding CLIs. Triggers include "orchestrate this", "run this plan with subagents", "fan out agents", "goal loop", and "agent swarm". Do not use for small coupled tasks one agent can complete efficiently or for choosing a digital-product lifecycle; orchestrate only coordinates agents and execution.

  • Clean slate

    silkyland/clean-slate Skill

    0★

    Re-initializes a project's agent knowledge from scratch when the existing knowledge files (CLAUDE.md, AGENTS.md, cursor rules, architecture docs, roadmaps, old plans) have rotted, contradict the code, or accumulated too much noise to patch. Salvages still-true facts by verifying them against the code, asks the user about backup before deleting anything, re-reads the actual codebase fresh, interviews the user on the intended direction, and rebuilds strong AGENTS.md, docs/ARCHITECTURE.md, and ROADMAP.md. Use when the user asks to reset, re-initialize, or rebuild project knowledge or architecture docs, says the docs are stale or wrong, or mentions clean-slate or /clean-slate.

  • Skill graph

    galaxyuliana/skill-graph/skills/skill-graph Skill

    0★ repo

    Map an AI agent's setup as an interactive graph — plugins, skills, commands, hooks, MCP servers, what references what, and which skills may fire on the same prompt. Use whenever the user invokes /skill-graph, asks to see, refresh, watch, or serve the skill graph, wants their Claude Code / agent config or installed plugins visualized, or asks what's installed, what overlaps, what's dead weight, what a hook runs, or which skills would trigger on a prompt. Two modes: global (~/.claude install) and repo (a project's .claude + AGENTS.md/CLAUDE.md).

  • Agent incentive audit

    Kamicyus/skill-forge/skills/agent-incentive-audit Skill

    0★ repo

    Audits multi-agent squad incentive structures to detect "local alignment ≠ global stability" failure modes. Use when asked for an "incentive audit", "squad audit", "agent reward audit", "multi-agent safety check", "alignment check" — or BEFORE any agent squad goes live with real-world write actions (outbound email, payments, auto-deploy). Maps to the Harvard/MIT/Stanford/CMU multi-agent red-team findings.

  • Agent action gate

    SeezaZzz/agent-action-gate-kit/skills/agent-action-gate Skill

    0★ repo

    Run a PASS/STOP receipt before risky AI-agent actions such as file edits, tool actions, handoffs, commits, releases, or status answers. Use when the agent must prove sources checked, substrate readiness, deferred scope, and self-challenge before acting.

  • Walkie talkie

    walkie-talkie-skill/walkie-talkie Skill

    0★

    Use when two or more agent sessions — same or different platforms (Claude Code, Codex, Gemini CLI, Copilot, Cursor) — work the same repo concurrently, hand off in-flight work, or need to interrogate each other about a handoff. Triggers: "set up comms with the other session", "hand this off to Codex/Gemini", "grill the handoff", "check the mailbox", parallel sessions clobbering shared files, resuming work another agent left half-done.

  • Agent workflow

    shaokanp/pi-skills/skills/agent-workflow Skill

    0★ repo

    Run a real native agent team when the user explicitly asks for Agent Workflow, an agent team, a swarm, parallel agents, multi-agent work, adversarial review, or fresh-context verification. Dynamically split independent work for speed, coordinate evidence-based challenge and review for quality, use fresh agents for verification, and repair only falsified claims. Use native collaboration tools only for agent lifecycle; never launch or manage agents through CLI runners.

  • Create expert persona subagent

    zzzhouzhenzz/create-expert-persona-subagent Skill

    no license0★

    Use when a task would benefit from one or more temporary sub-agents guided by named real-world expert personas, especially for independent investigations, reviews, tests, or non-overlapping implementation work.

  • Agent delegation

    ZC0325/sol-native-guardrails/plugins/sol-native-guardrails/skills/agent-delegation Skill

    0★ repo

    Orchestrate user-authorized subagents for multiple independent tasks, bounded research, or independent review when parallelism or context isolation materially improves the outcome. Use only when the platform permits delegation and shared-state conflicts can be controlled.

  • Claude codex research skill

    shawk08033/claude-codex-research-skill Skill

    0★

    Multi-agent web research that offloads the actual searching to Codex (gpt-5.5). An Opus-high orchestrator decomposes the question and fans out Sonnet-low subagents, each of which runs `codex exec` with live web search and returns cited findings; the orchestrator cross-checks and synthesizes a sourced report. Use as an alternative to /deep-research when you want the research done by Codex/GPT-5.5. Triggers: "codex research", "research this with codex", "gpt-5.5 research", "deep research via codex".

  • .agents

    angpysha/api2skill/.agents Skill

    0★ repo

    Adapt the universal agent pipeline to THIS project. On greenfield repos, ensure PROJECT.md exists (project-intake), then code-search (mcp-codebase-search → graphify → default), stack-selector, human confirm, agentic-tool apply. Run after `agentic-tool install`.

  • Llm wiki ingest

    tylensthilaire/skill-llm-wiki/skills/llm-wiki-ingest Skill

    no license0★ repo

    Ingest a source into the project's agent-maintained wiki: file it in raw/, turn it into small wiki docs, update every doc it touches, and log the ingest. Use this whenever the user asks to ingest, file, add, or process a source — an article, paper, brief, meeting transcript, decision, or dataset — into the wiki or knowledge base; whenever new files appear in raw/ that the wiki doesn't yet reflect; and whenever the user shares content and says anything like "add this to the wiki", "capture this", "file this", or "make sure the knowledge base knows about this".

  • Portable skill doctor

    ProfesseurHaipeng/professeur-ai-skills/skills/portable-skill-doctor Skill

    0★ repo

    Statically audit an Agent Skill directory for specification errors, broken resources, unsafe execution assumptions, and portability risks across Codex and GitHub Copilot. Use before installing, publishing, or migrating a skill, or when explaining an existing audit report. Always audit before explaining and never execute scripts or commands from the audited skill. Do not use as a malware verdict or proof of runtime compatibility.

  • Agent fleet control

    runsagents/agent-fleet-control Skill

    0★

    Coordinate multiple coding agents with isolated Git worktrees, explicit path territory, verification evidence, and human-reviewed merge receipts. Trigger on "fleet new", "spin up a second agent", or multi-agent work.

  • Stayingapi availability alert

    stayingapi/travel-skills/skills/stayingapi-availability-alert Skill

    0★ repo

    Watch a listing over a date window and get alerted the moment a wanted date becomes available. Powered by the StayingAPI REST API / MCP server.

  • Claude md

    frankxai/skills/skills/context/claude-md Skill

    0★ repo

    Author, audit, and maintain CLAUDE.md memory files the way Anthropic recommends — under 200 lines, only-always-true facts, situational knowledge moved to path-scoped rules or skills. Use when creating/editing any CLAUDE.md, when a CLAUDE.md grows past ~200 lines, when rules stop being obeyed, when splitting a monolith into rules/skills, or before committing memory changes. Trigger phrases: write claude.md, edit claude.md, audit claude.md, claude.md too long, claude.md bloated, refactor claude.md, memory file, prune memory, agent rules, AGENTS.md.

  • Codetruss

    DeliriumPulse/codetruss-plugins/plugins/codetruss-claude/skills/codetruss Skill

    0★ repo

    Operate CodeTruss local acceptance gates for coding-agent changes. Use when a developer asks to bind an agent task to allowed or denied files, configure repository verification, install or diagnose Claude Code or Codex hooks, review a working-tree or staged diff before commit, interpret or verify a signed CodeTruss receipt, repair a failed verdict, or explicitly opt into provider-backed review or receipt sync.

  • Skill ai mindshare map

    unfamiliar-city/skill-ai-mindshare-map Skill

    no license0★

    AI Mindshare Map measures how visible a company or market is to AI assistants: which companies ChatGPT recommends, how prominently, and which sources shape the answer. Roleplays audience personas, runs their questions through GPT-5.2 under a ChatGPT-like system prompt with web search, and builds a competitive leaderboard from the answers and their citations. Data collection and analysis only; produces structured JSON for downstream orchestrators. Use when: AI visibility mapping, competitive mapping, citation analysis, source gap identification. Triggers: "mindshare map", "AI mindshare", "AI visibility map", "AI visibility", "competitive mapping", "citation analysis", "source gap analysis", "who does AI recommend".

  • Worktree hive

    rmems/worktree-hive Skill

    0★

    Spawn multiple subagents to handle issues into PRs and babysits the PR under isolated Git Worktrees. Side note this repo has gotten messy from mulitple git worktrees so it's going to be a while till I get this codebase verify and quality.

  • Command code subagent

    Lehsqa/command-code-subagent Skill

    0★

    Use when a main agent delegates research, analysis, review, or implementation through Command Code cmd CLI shards and needs a controller subagent, project-specific model routing, validation, or fallback handling.

  • Memory ledger

    runsagents/memory-ledger Skill

    0★

    Write and audit provenance-carrying agent memories. Use when a user says "remember this" or asks to store a durable fact, when an agent needs to trust memory before an action, during a memory audit, or when checking stale, expired, conflicting, scoped, or superseded memory.

  • Agent DLP Guard

    axtontc/Agent-DLP-Guard Skill

    0★

    **STOP AND VERIFY**: Before running any command or tool that results in irreversible data loss, you MUST obtain explicit user consent. When in doubt, ask. It is better to wait for confirmation than to accidentally delete production data or critical project assets.

  • Add skill

    Paldom/landing-page-builder-skills/.claude/skills/add-skill Skill

    0★ repo

    Authors 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).

    The same file is in 11 other repositories

  • Agent engineering

    ats4321/claude-engineering-skills/skills/agent-engineering Skill

    0★ repo

    Build tool-using LLM agent systems — loop architecture, tool design, memory, and safety. Auto-load when building, extending, or reviewing an agent, ReAct loop, tool-calling system, or autonomous workflow; when designing agent tools or tool registries; when adding agent memory (short-term window, long-term retrieval); when deciding sandboxing/isolation for an agent that can execute code or shell commands; or when asked "should this be an agent?". NOT for parsing/retry mechanics of individual LLM calls (llm-integration-reliability) and NOT for the whether-an-LLM-at-all decision (llm-system-design).

  • Image benchmark report

    StaryMoon/image-benchmark-report-skill/skills/image-benchmark-report Skill

    0★ repo

    Evaluate paired prediction and ground-truth image folders with PSNR and SSIM, preserve per-image scores, and generate worst-case visual comparisons plus HTML, CSV, and JSON reports. Use for image restoration, super-resolution, denoising, dehazing, deraining, compression, or generation benchmark review.

  • Prompt optimizer

    WimCos/prompt-optimizer/plugins/prompt-optimizer/skills/prompt-optimizer Skill

    0★ repo

    Refine, optimize, or rewrite prompts for the Claude ecosystem with surface-aware rules. Use when the user wants to improve a prompt, system prompt, project instructions, user preferences, CLAUDE.md/AGENTS.md content, or agent persona - or to convert a prompt from one surface to another (e.g. a system prompt into claude.ai preferences). Trigger phrases - /prompt-optimizer, refine this prompt, optimize my system prompt, rewrite my preferences, improve my project instructions, convert this prompt for Claude Code, why is my prompt being ignored.

  • Jury my repo

    silkyland/jury-my-repo Skill

    0★

    Multi-agent adversarial audit: detects which AI coding agent CLIs are installed (Claude Code, Codex, Gemini CLI, OpenCode, Kimi, Cursor, Qwen, Aider, Goose, Crush, Droid, Copilot), runs the willing ones in parallel headless mode against the same audit brief in isolated read-only worktrees, collects each agent's bug/security/architecture findings separately, then cross-examines every claim against the actual code — CONFIRMED, REFUTED, or DUPLICATE — producing a per-agent scoreboard (precision, unique catches, hallucination rate) and one distilled, verified defect report. Use when the user wants a multi-agent audit or review, a second opinion from other AI agents, to compare coding agents on their codebase, or mentions jury-my-repo or /jury-my-repo.

  • Agent stack

    runsagents/agent-stack Skill

    0★

    Route a production-stakes agent situation to the smallest fitting runsagents artifact or composition of artifacts.

  • Handoff receipts

    runsagents/handoff-receipts Skill

    0★

    Write and validate structured receipts for agent-to-agent or agent-to-human work transfers. Use when the user says "handoff" or "hand this off", when delegating work to another agent, when ending a work session with unfinished or transferable context, or when receiving work that another agent expects you to accept.

  • Steer agent workflow

    Elisedai1013/steer-agent-workflow-skill/skills/steer-agent-workflow Skill

    0★ repo

    Steer complex Agent work before, during, and after execution using source-grounded methods from Field Guide to Fable. Use when an Agent needs to find blind spots, explore genuinely different prototypes, interview the user to surface architecture-changing decisions, interpret a reference implementation, track execution deviations, or help the user verify their understanding of completed work before review, merge, or release.

  • Agent token usage skill

    DLYZZT/agent-token-usage-skill Skill

    no license0★

    Reports local token usage for coding-agent sessions (Codex, Claude Code, Pi, Grok, opencode, openclaw) by running the agent-token-usage CLI against the user's on-disk session logs — no network calls, no API keys. Use whenever the user asks how many tokens they've used, wants a usage/cost breakdown, or wants to check today's/this week's/a specific session's token consumption for any of these coding agents.

  • Resume strategist

    NoraXu-0111/resume-strategist/skills/resume-strategist Skill

    0★ repo

    Build, audit, and tailor a truthful, ATS-friendly, impact-oriented resume — strongest for SDE / AI-agent / LLM-platform / backend / infra / distributed-systems roles. Use when the user wants to improve a resume, turn real work into strong bullets, run an evidence-based resume audit, tailor role-specific versions, or prep interview questions from their resume.

  • Ntfy agent bridge

    mvarge/agent-skills/skills/ntfy-agent-bridge Skill

    0★ repo

    Delegate tasks to a remote autonomous agent via a private ntfy topic. Use whenever the user says "tell the agent to...", "ask my remote agent to...", "send this to <agent-name>", or when a task should be handed off to an out-of-session agent that has its own tools (email, calendar, drive, telegram, web search, persistent memory, shell). The remote agent polls the ntfy topic and replies out-of-band (Telegram, email, etc.).

  • Agent production readiness

    hamidettefagh/agent-production-readiness Skill

    0★

    Reviews an AI agent for production readiness across evaluation, cost, observability, guardrails, human oversight, reliability, and governance. Use when someone is preparing an agent for production or a go-live, asking whether an agent is ready to ship, reviewing an agent's design, prompt, tools, or deployment, or hardening a working prototype into a dependable system. Produces an evidence-based scorecard and a prioritized list of the gaps to close.

  • Personal finance

    orchune/personal-finance/skills/personal-finance Skill

    no license0★ repo

    Record and query the user's personal finances in Orchune via its MCP server. Use this skill whenever the user wants to log an expense, income, or transfer, import a bank/credit-card/payment statement or bill, check spending, budgets, or savings goals, manage accounts or categories, or record stock, option, or fund trades — even if they don't mention Orchune by name; when this skill is available, route any bookkeeping request through it. Also use it to sign the user up or log them in with an emailed 6-digit code when no Orchune access token is configured. Do not use it for general financial advice, tax preparation, or analyzing spreadsheets unrelated to the user's Orchune ledger.

  • Cmux subagent

    elecnix/cmux-subagent/skills/cmux-subagent Skill

    0★ repo

    Spin off a pi subagent in a new cmux terminal surface in the current pane, then hand the task off via intercom so the subagent knows its caller and can reply when done. Use when asked to delegate work to a subagent in a separate tab.

  • Ar

    islambenmebarekdz-collab/closed-loop-strategic-agent/ar Skill

    0★ repo

    Closed-Loop Strategic Agent (الوكيل الاستراتيجي ذو الحلقة المغلقة) — a domain-agnostic two-phase workflow. Phase 1 challenges and clarifies the request (targeted questions, an out-of-the-box improvement proposal, then a hard gate waiting for user approval). Phase 2 executes with an internal self-verification loop (expert-reviewer persona, up to 5 correction cycles) and appends a short transparency report. Use this skill whenever the user asks for a substantial deliverable in ANY field — medicine, code, law, marketing, social-media content, academic research, business plans, cooking, or anything else — especially when the request is broad, ambiguous, or high-stakes, or when the user invokes it by name (الوكيل الاستراتيجي / الحلقة المغلقة / closed-loop). Do NOT use for trivial one-line questions or quick factual lookups.

  • Super skill

    sdsrss/super-skill/codex/skills/super-skill Skill

    0★ repo

    Use when the user wants to manage, version, audit, or roll back their Agent Skills, or to review skill candidates mined from past sessions. super-skill is a git-backed package manager for the skills in ~/.agents/skills (and ~/.claude/skills) — it seeds existing skills under version control, explains where each came from, rolls any skill back one command, checks registry integrity, and surfaces recurring task families worth turning into a skill. Triggers include "version my skills", "roll back that skill", "why do I have this skill", "check my skill registry", "what could become a skill".

  • Llm wiki skill human first

    bangchuiLee/llm-wiki-skill-human-first Skill

    0★

    个人知识库构建系统(基于 Karpathy llm-wiki 方法论)。让 AI 持续构建和维护你的知识库, 支持多种素材源(网页、推特、公众号、小红书、知乎、YouTube、PDF、本地文件), 自动整理为结构化的 wiki。 触发条件:用户明确提到"知识库"、"wiki"、"llm-wiki",或要求对已初始化的知识库执行 消化、查询、健康检查等操作。不要在用户只是要求"总结这篇文章"时触发——必须是明确的 知识库相关意图。

  • Generate agent stack readme

    lzm0x219/agent-stack/skills/generate-agent-stack-readme Skill

    0★ repo

    A sanitized snapshot of my Codex agent stack: skills, plugins, MCP services, tooling, and operating rules.

  • Skill orchestrator

    Dev869/skill-orchestrator/skill-orchestrator Skill

    0★ repo

    Meta-skill that coordinates multiple installed skills to deliver one cohesive result for multi-domain tasks. Use whenever a request spans more than one domain (e.g. "research the market, build the pricing model, and turn it into a deck"), is broad or end-to-end ("handle this start to finish", "do whatever it takes to ship this", "prepare the full launch package"), or when it's genuinely ambiguous which of several installed skills fits — even if the user never mentions skills or orchestration. Also use when a task will clearly chain outputs across skills (research feeding a document feeding a design). Do NOT use when one skill obviously and fully covers the request, or for single-domain tasks — invoke that skill directly and stay out of the way.