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

  • Update stock mcp

    mifochen/update-stock-mcp/update-stock-mcp Skill

    no license0★ repo

    UpdateStock MCP 服务技能 —— 通过 stdio 模式调用 UpdateStock 脚本, 提供 A 股 DuckDB 数据库管理功能:创建数据库、全量/增量更新股票数据、查询股票行情。 设计的数据库方案兼容 QuantAll(全A解析)计算引擎。 触发条件:用户提到"UpdateStock"、"创建股票数据库"、"更新股票数据"、"获取股票数据"等。

    The same file is in 2 other repositories

  • Create my team

    silkyland/create-my-team Skill

    0★

    General-purpose multi-agent teamwork: for any sizable task, probes what subagent-spawning capability the current platform actually exposes (Claude Code agents/workflows, Codex multi-agent tools, OpenCode subagents — or none, with an honest sequential fallback), designs a team topology sized to the task (scout swarm, pipeline, panel-and-judge, manager-worker), sends every member a standardized mission brief, researches BEFORE executing, presents the synthesized plan at a user gate, then spawns the execution team and cross-checks every deliverable at an integration gate before merging. Use when the user wants to spawn agents or subagents, parallelize research or work across a team, says "use a team/multi-agent" for a task, or mentions create-my-team or /create-my-team.

  • Llm wiki vault manager

    chang769/llm-wiki-vault-manager Skill

    no license0★

    Manage source-backed LLM-wiki and Obsidian-style Markdown vaults as learning databases for humans and AI agents. Use when an AI coding agent such as Codex or Claude Code needs to scaffold a raw/wiki/schema knowledge database, organize provided links or files into the proper raw material directory, ingest raw sources into durable wiki notes, update the vault structure/index/log, help the user and agent learn a chosen area, answer vault questions with citations, lint indexes/backlinks/logs, automatically review, summarize, commit, and push vault enrichment changes with git, download or preserve raw material, or distill repeated source-backed workflows into reusable agent skills.

  • Agent fabric

    jroell/agent-fabric/templates/skills/agent-fabric Skill

    0★ repo

    Operate, audit, repair, and extend the agent-fabric on this machine — the canonical hub (~/.agent-fabric) that shares skills, rules/instructions, and memory across ALL agent harnesses (Claude Code, Codex, Warp, Cursor, Gemini CLI, Grok CLI, OpenCode, Hermes). Use when the user asks to "audit my agent setup", "sync my agents", "add this skill/rule to all agents", "install a skill everywhere", "wire up a new agent", "share skills between agents", mentions agent-fabric / canonical skills / shared memory, or when a skill or agent config seems broken after a migration or install. ALSO USE BEFORE running any migration, conversion, or bulk-install tool that writes to skill directories or agent config files, and AFTER such a tool runs.

  • Agentic change audit

    landco-llc/agentic-change-audit Skill

    0★

    Audit software changes, pull requests, local diffs, or release candidates for target identity, scope, correctness, regression, security, evidence quality, and merge or release readiness. Use for independent audits, fixed-HEAD reviews, focused re-audits, docs-only audits, or verdicts such as PASS, CHANGES REQUESTED, BLOCKED, and NOT AUDITABLE. Do not use to implement fixes or perform a generic style-only review.

  • Design gate

    hamidettefagh/two-gates/skills/design-gate Skill

    0★ repo

    Decides the architecture for an AI feature before it is built, whether the work should be a deterministic workflow, a workflow with a model at the judgment points, a single model call, a single agent with tools, or multi-agent. Use when someone is scoping or designing a new AI use case, asking whether it needs an agent, reviewing a PRD, ticket, or spec for an agent, or pushing back on a design that reaches for an agent where a workflow would do. Also covers Salesforce Agentforce designs, deciding whether a use case should be a Flow, a prompt template, an Agentforce topic with actions, or multiple agents. Reads the design input, extracts the deciding signals with evidence, runs a deterministic verdict, and writes an architecture decision record that biases toward the simplest thing that works.

  • Agentic system design

    sandev/engineering-os/skills/ai-and-agentic-systems/agentic-system-design Skill

    0★ repo

    Design reliable LLM and agent-based systems, including tool design, context management, evaluation, guardrails, and failure handling. Use when building an AI agent, LLM feature, RAG pipeline, or agentic workflow, or when the user asks how to make an LLM system reliable, testable, or production-ready.

  • Jtbd agent builder

    dstevenson1981/jtbd-agent-builder Skill

    0★

    Build agents that replace jobs, starting from a one-sentence business outcome. Use when a user wants to create an agent or skill, automate a job-to-be-done, achieve an outcome ("more sales through digital channels", "cut support cost", "hire faster"), figure out which jobs an agent should do for their business, or set up a self-improving agent with evals. Diagnoses the user's actual system from real data, finds the binding constraint, proposes a dollarized agent portfolio, and emits runnable skills for Claude Code or Codex with metric contracts, graduated autonomy, and a PR-based improvement loop.

  • Agentic critique

    Thecsiz/agentic-principles/skills/agentic-critique Skill

    no license0★ repo

    Evaluates, re-storyboards, and designs agentic AI experiences where an agent does work and a human approves, steers, or supervises it. Use for requests such as 'critique this agent flow', 'is this actually agentic', 'is the approval real', 'show me the agent-led version', or 'help me design an agentic flow', and when a user shares a storyboard, screenshot, Figma URL, prototype, or description of an AI feature. Runs two required diagnostics—who is doing the work, and whether approval is meaningful—then scores P0–P15 and checks 13 named anti-patterns. Evaluates agency, oversight, trust, consent, and autonomy; it does not replace usability, accessibility, visual-design, security, legal, or compliance review.

  • Prompt architect

    chuaitry-edu/try-skills/hermes/prompt-architect Skill

    no license0★ repo

    Prompt生成与优化助手。认为大多数烂Prompt不是缺技巧,是没说清楚边界。属于try学习体系的元工具层。

  • Kb cli

    JakePaterson/growth-garage-toolkit/skills/kb-cli Skill

    0★ repo

    Use when managing an agent-facing knowledge base — keeping docs fresh, checking for stale or low-confidence articles, compiling one wiki source into every IDE's config file (CLAUDE.md, AGENTS.md, GEMINI.md, Cursor/Cline rules), regenerating the article index, scaffolding a new project wiki, searching wiki articles by confidence, or filing new sources into a wiki. Trigger words include "kb search", "kb lint", "kb compile", "kb index", "kb init", "knowledge base", "wiki health", "stale docs", "update the wiki", "compile IDE config", "new project wiki".

  • Frontrun deal flow

    jongall45/frontrun-skills/skills/frontrun-deal-flow Skill

    0★ repo

    Source startups before their round using Frontrun: live pre-round deal signal from the follow graphs of 1,000+ venture investors on X. Use when asked to find early-stage startups, source deals, run a thesis search, check which companies investors are converging on, diligence a startup's momentum, or build a deal-flow pipeline or agent. Works via the Frontrun MCP server or REST API.

  • Open health agent

    w2478328197-arch/open-health-agent/skills/open-health-agent Skill

    0★ repo

    Use for private personal health and fitness logging, wearable or Google Health/ghealth sync, Excel health ledgers, WeChat text/voice/photo intake, food and nutrient estimates, calorie or TEF calculations, exercise/recovery advice, and persistent health goals. Except when urgent emergency direction must come first, explain the data flow, capabilities, privacy exposure, uncertainty, and non-medical boundary in the user's language on the first use in every conversation.

  • Prompt sharpener

    mattrmartin/prompt-sharpener/prompt-sharpener Skill

    0★ repo

    Turn a rough, half-formed prompt into a sharpened, high-performing one using Anthropic's own prompt-engineering evidence. Use this whenever the user hands over a draft prompt, pastes a prompt and asks for feedback, describes a task they want to prompt Claude (or another model) to do, or says things like "help me write a prompt", "improve/fix/tighten this prompt", "make this prompt better", "sharpen this", "turn this into a system prompt", or "why isn't this prompt working". Also use when the user is building a system prompt for an app, agent, custom assistant, or GPT. Handles both one-off chat/task prompts and reusable system prompts. Reach for this even when the user doesn't say the word "prompt" but is clearly drafting instructions for a model to follow.

  • Token firewall team

    WdBlink/token-firewall-team/skills/token-firewall-team Skill

    0★ repo

    Run Codex Agent Teams through the native subagent lifecycle while preserving bounded contracts, Git truth, deterministic validation, independent verification, and blind final review. Use an external CLI Adapter only when the user explicitly requests a third-party platform such as Claude, MiniMax, or OpenCode.

  • Sdd skill

    JMSolorzano-13/sdd-skill Skill

    0★

    Specification-Driven Development (SDD) orchestration skill. Use this skill whenever the user mentions .sdd-skill/, state.md, @planner, @architect, @implementer, @review, @tester, @sentinel, SDD, /sdd-skill commands, spec.md in a project context, or wants to continue development across IDEs (Cursor, Claude Code, Antigravity, Open Code). Also trigger when the user wants to start a new project with agentic methodology, adopt SDD in an existing project, manage features/hotfixes with agent roles, or resume work after switching tools. This skill governs a polymorphic agent system where a single AI assumes different roles based on filesystem state — never chat history. Trigger even if the user just says "continue", "resume project", or "what's next" and a .sdd-skill/ folder is likely present.

  • Eval authoring

    soltraveler-sri/Steerable-Protocol/skills/eval-authoring Skill

    0★ repo

    Author deterministic Steerable eval fixtures for a target integration by deriving cases from its capability registry, validating them against the eval schemas, running them through the target adapter, and emitting coverage plus duplicate-avoidance summaries.

  • Pgbeam connect

    sferarc/pgbeam-skills/pgbeam-connect Skill

    no license0★ repo

    Wire an AI agent to a Postgres database safely with PgBeam. Use this when a task needs the agent to read or query a real Postgres database and you want scoped, read-only-by-default, PII-masked, budgeted, audited, revocable access instead of handing the agent a full-privilege connection string. Covers getting a scoped credential, choosing between a guarded connection string and the hosted MCP endpoint, and paste-ready Claude Code, Cursor, and VS Code config.

  • Agent or workflow

    hamidettefagh/agent-or-workflow Skill

    0★

    Decides the architecture for an AI feature before it is built, whether the work should be a deterministic workflow, a workflow with a model at the judgment points, a single model call, a single agent with tools, or multi-agent. Use when someone is scoping or designing a new AI use case, asking whether it needs an agent, reviewing a PRD, ticket, or spec for an agent, or pushing back on a design that reaches for an agent where a workflow would do. Also covers Salesforce Agentforce designs, deciding whether a use case should be a Flow, a prompt template, an Agentforce topic with actions, or multiple agents. Reads the design input, extracts the deciding signals with evidence, runs a deterministic verdict, and writes an architecture decision record that biases toward the simplest thing that works.

  • Pm trace eval

    Nyota-tree/pm-trace-eval/skills/pm-trace-eval Skill

    0★ repo

    Evaluates AI agent conversation traces for product managers and produces PM-ready reports with user-defined north-star metrics, 1-5 rubric scores, red/yellow/green signals, and per-turn good/bad original quotes. Use when the user asks to review, score, or audit a dialogue trace, conversation log, persona eval, agent chat eval, LangSmith export, eval transcript, multi-turn assistant quality, or PM review of agent logs (Markdown, JSON, or pasted text). Do NOT use for code review, prompt editing, LangGraph or engineering root-cause debugging, or modifying agent configuration.

  • Knowledge base awareness

    kpince/pi-knowledge-cognition/skills/knowledge-base-awareness Skill

    0★ repo

    Recognize when repeated explanation, cross-agent work, durable corrections, changing evidence, or consequential decisions may warrant a knowledge base. Use before creating storage, schemas, dashboards, or automatic memory; determine whether durable governed knowledge is actually needed.

  • Agent browser

    DrOlu/agent-skills/skills/agent-browser Skill

    no license0★ repo

    Automates browser interactions for web testing, form filling, screenshots, and data extraction. Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information from web pages.

  • Agents md init

    K95M65/AI_ONBOARD/skills/agents-md-init Skill

    0★ repo

    Bootstraps an AGENTS.md file for a project by detecting its stack, package manager, and build/test/lint commands. Use when a repo has no AGENTS.md (or CLAUDE.md) yet and the user wants to onboard AI tools, or asks to create/generate agent instructions for the project.

  • Auto optimize codex agents md

    ChanTinPing/auto-optimize-codex-agents-md Skill

    0★

    Mine local Codex active and archived session history for durable user corrections, dissatisfaction, quality standards, communication preferences, and safety boundaries; then suggest or explicitly auto-apply scoped changes to project-root or global AGENTS.md managed blocks. Use when asked to optimize Codex behavior memory, review or withdraw learned AGENTS.md rules, run incremental daily AGENTS memory maintenance, or configure that maintenance as a scheduled task. Do not use to mine workflow Skills or rewrite unowned AGENTS.md content.

  • Agent duel

    edycutjong/agentduel/skills/agent-duel Skill

    0★ repo

    Field an autonomous duelist in AgentDuel — a World Cup duel arena on Injective EVM where two agents stake opposing sides of a fixture over x402 and the winner is paid on-chain. Use when the user wants to enter a duel, take a side on a match, stake a pick that can't be faked, or run a model-vs-model wager. Handles: reading a duel card, choosing a side (bring your own signal), paying the 0.10 USDC x402 entry (the receipt is the pre-kickoff commitment), polling for settlement, and reporting the payout. Harness-agnostic (Claude Code / Cursor / Codex).

  • Markdown idempotent section merge

    h8nc4y/markdown-idempotent-section-merge Skill

    0★

    Idempotently replace-or-append one Markdown section (a "## Heading" block) without corrupting code fences. Use on symptoms like: an automated section merge duplicated the section on every run, a replace range was cut short at a "##" line inside a code fence, a heading inside a code fence was treated as the next section boundary, a "###" subheading ended the section early, a README / AGENTS.md / CLAUDE.md updater never converges (the second run still changes the file), or a leftover fence swallowed the rest of the document after a bad merge. Covers fence-aware heading scanning, fixed begin/end marker boundaries, the single-H2 invariant (^##[^#]), and an apply-twice-diff-zero verification recipe with a tested reference implementation.

  • Agentic interface foundation

    nick3/vibe-design-skills/skills/agentic-interface-foundation Skill

    0★ repo

    Define the product-level experience and rendering contract for AI assistants, agents, copilots, generative UI, adaptive workflows, human-in-the-loop decisions, and structured Agent-to-UI messages. Use this skill whenever a product includes conversational or autonomous actions, dynamically generated interface regions, agent progress, uncertainty, authorization, generated reports, or requests mentioning AX, GenUI, A2UI, chat UI, copilot, or agentic experience. Do not use it to make stable settings, navigation, or dashboards generative without a contextual need.

  • Openagent workspace inspect

    JoinOpenAgent/openagent-skills/skills/openagent-workspace-inspect Skill

    0★ repo

    Read-only inspection of an OpenAgent workspace: search conversations, fetch transcripts, look up visitors, inspect the AI's MCP audit trail, list configured agents / workflows / KB documents. Powered by OpenAgent's MCP server.

  • Realistic video prompting

    HiAPIAI/realistic-video-prompting Skill

    0★

    Design and review model-agnostic video prompts that should resemble captured real-world footage, including phone clips, consumer camcorder recordings, analog tape, small-gauge film, action cameras, security cameras, webcams, observational documentary, and found footage. Use only when the user asks for natural capture, physical camera behavior, documentary evidence, or reduced AI polish. Do not trigger for ordinary video generation or polished commercial video requests.

  • Pdf to editable word

    michellema8890/pdf-to-editable-word-skill/skills/pdf-to-editable-word Skill

    0★ repo

    Convert PDF to Word or PDF to DOCX while preserving the original page layout and keeping text editable, searchable, and selectable. Use for PDF-to-Word conversion, editable DOCX creation, fixed-layout documents, table-heavy PDFs, vertical text, resumable long-document conversion, or output validation. Works as a portable Agent Skill for Codex, Claude Code, and other agents that can run the CLI. Do not use for scanned PDFs without a text layer unless OCR is available.

    The same file is in 1 other repository

  • Agent project quality audit

    Wang-Yeah623/skills/agent-project-quality-audit Skill

    no license0★ repo

    Audit software-project quality through repository mapping, risk-based scope selection, evidence-backed defect confirmation, test-first minimal repair, layered regression checks, and persistent `.quality` records. Use when asked to understand an unfamiliar repository, establish acceptance/risk/coverage artifacts, run a bounded read-only or fix-enabled quality audit, investigate a runtime failure, verify a previous fix, continue an audit without repeating covered areas, or assess release convergence. Do not present this as a penetration test, compliance audit, or substitute for a dedicated security review.

  • Agentic loop

    m83iyer/agentic-loop-skill/agentic-loop Skill

    0★ repo

    Turn a substantial multi-step objective into a bounded, resumable execution loop with frozen acceptance criteria, one active task, retry limits, and receipt-backed completion. Use when the user asks an agent to keep working through milestones, continue autonomously, resume after interruption, or prove completion. Do not use for simple one-step tasks, open-ended monitoring without a host scheduler, or work that would expand permissions or external-action authority.

  • Engineer agent loops

    Rowan2005321/loop-contracts/skills/engineer-agent-loops Skill

    0★ repo

    Design, audit, and repair bounded agent execution loops by connecting harness capabilities to explicit state, progress signals, budgets, verification, recovery, observability, human gates, and stop conditions. Use when Codex needs to turn a recurring or long-running agent workflow into a machine-checkable loop contract, diagnose retries or false completion, compare loop designs, or add safety and resumability to autonomous work.

  • Route github agent decisions

    cristianmoroaica/bountyverdict-mcp-skill/skills/route-github-agent-decisions Skill

    0★ repo

    Diagnose why a GitHub Actions run failed and find its root cause; decide whether to retry that failed Action once; check or rank GitHub bounties; audit AGENTS.md readiness; detect MCP schema drift.

  • Agent fix loop

    kadenn/skills/skills/agent-fix-loop Skill

    0★ repo

    Run a measured reliability loop for LLM-agent and model-driven features. Use when the user asks to harden an agent, improve eval reliability, diagnose judge-versus-trace disagreement, reduce flaky tool use, or autonomously repeat fix-test-measure cycles toward a defined target. Establish a trace-verified baseline, isolate one failure mode, make one minimal change, validate it, compare repeated trials, keep or revert, and stop at the agreed threshold. Do not use for a one-off deterministic bug that does not require behavioral measurement.

  • Context diet skill

    cskwork/context-diet-skill Skill

    0★

    Cut Claude Code's per-request system-prompt bloat by measuring and disabling unused tools, bundled skills, and features via settings.json. Use when the user wants to reduce Claude Code token usage or context size, shrink the system prompt, make requests cheaper or faster, run a logging proxy to rank tools by size, or asks about /context, disableBundledSkills, disableWorkflows, or permissions.deny for tool definitions.

  • Goal chunking agent architect

    halans/goal-chunking-agent-architect Skill

    no license0★

    Design and generate a new AI agent (or agentic workflow) that reliably reaches a specific goal. Use this when the user asks to create, design, architect, scaffold, or "spin up" an agent or subagent aimed at a concrete outcome — and especially when they mention goal chunking, the Harada Method, mandala/OW64 goal decomposition, or want a rigorous, self-reliant agent instead of an ad-hoc prompt.

  • Seo geo aso agent skill

    artificialguybr/seo-geo-aso-agent-skill Skill

    0★

    Optimize websites, web pages, and mobile apps for organic visibility across all three modern discovery channels — SEO (Google/Bing organic search), GEO/AEO (getting cited by AI answer engines: Google AI Overviews & AI Mode, ChatGPT, Perplexity, Gemini, Claude, Copilot), and ASO (Apple App Store + Google Play). Use this skill WHENEVER the user wants to improve search rankings, get found by AI/LLMs, increase app downloads, or asks to audit, optimize, or fix anything about a site's or app's discoverability — including adjacent phrasings like "rank higher", "show up in ChatGPT", "AI search visibility", "Core Web Vitals", "structured data / schema", "keywords", "meta tags", "robots.txt for AI crawlers", "llms.txt", "app store listing", "app keywords", "screenshots/conversion rate", or "why isn't my page/app getting traffic". Runs entirely on FREE tools — never requires paid subscriptions. Prefer this skill even when the user doesn't say "SEO/GEO/ASO" explicitly.

  • Agent runbook lint skill

    rogerchappel/agent-runbook-lint-skill Skill

    0★

    Local-first lint checks for agent automation runbooks

  • Model router

    Renga154/model-router Skill

    0★

    Route AI development tasks to the optimal model and reasoning effort by difficulty. Use when starting any coding/implementation/debugging task to decide which model tier should execute it, when the user asks to optimize model usage or cost, or mentions model routing, モデル振り分け, 難易度判定, or コスト最適化. Works on Claude Code, Codex, and other Agent Skills-compatible platforms.

  • Add skill

    Paldom/mcp-server-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

  • Add skill

    Paldom/screenshooter/.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

  • Prepare project handoff

    TunaSaiko/prepare-project-handoff/plugins/prepare-project-handoff/skills/prepare-project-handoff Skill

    0★ repo

    为长会话或跨 Session 工作按当前决定渐进恢复可验证的项目判断能力:只读定位项目与权威来源、核对 Git 和工作区、区分导航/讨论/调查/执行边界/当前裁决,并在需要时生成安全的下一轮接手提示词。用于用户说 handoff、交接、ctx/context 快满、换 session、新会话接手、继续未完成项目、查看当前状态,或需要初始化、校验 `.handoff/authority.json` 时。

  • Multi agent

    zhongmaomao/multi-agent/skills/multi-agent Skill

    0★ repo

    Decompose an authorized objective or implementation brief into dependency-aware tasks, delegate independent work to subagents, coordinate shared files and interfaces, integrate results, and run independent verification. Use when a user explicitly asks for multi-agent work, parallel agents, subagents, delegation, worker and reviewer teams, or execution of a task plan through multiple agents. Do not invoke merely because a task has multiple steps.

  • Codex first

    oxfern/codex-first Skill

    no license0★

    Claude Code-only implementation routing: delegate implementation work to Codex CLI running GPT-5.6-sol at high reasoning effort while Claude specifies, reviews, and verifies. Use only when the active agent is Claude Code. Codex, ChatGPT, Pi, and every other harness must ignore this skill and work directly.

  • Repo reentry

    zheyuanhu2-sketch/repo-reentry/repo-reentry Skill

    0★ repo

    Create, verify, switch, and resume durable, agent-neutral repository memory so work can move across any AI coding agent or fresh session without the user repeating context. Use when the user wants to hand off a coding project, freely switch between Claude Code, Codex, Hermes Agent, Cursor, Copilot, Windsurf, Cline, OpenCode, or another agent, preserve long-lived project knowledge and current task state in project files, recover after context loss, validate whether a saved checkpoint is stale, create an AI handoff or reactivation prompt, or connect agent instruction files to the latest continuation state.

  • Audit agent harness

    cristianmoroaica/bountyverdict/skills/audit-agent-harness Skill

    0★ repo

    Audit public GitHub repositories with HarnessVerdict before autonomous coding. Use when checking AGENTS.md, CLAUDE.md, GEMINI.md, Copilot or Cursor instructions, validating SKILL.md files, investigating ignored or stale guidance, or deciding whether a repository is structurally ready for coding agents.

  • Graph

    NicholasSpisak/graph/skills/graph Skill

    0★ repo

    Drafts an agent graph — the map of nodes, checkpoints, and routes (a state machine) that a repeatable job travels — from a job description or a stated outcome. Compiles the map to a schema-validated lock file, renders a visual review surface the user must approve before any run, and emits multi-model drivers (Claude Code Workflow + chained Codex CLI). Use when the user says "build a graph", "/graph", "map this pipeline", "turn this outcome into a pipeline", "design the topology", "wire nodes and checkpoints", wants validation gates between agent steps, or asks whether a job needs a loop or a graph.

  • Orchestrate olympus

    dm1681/skills/skills/orchestrate-olympus Skill

    no license0★ repo

    Create, resume, pause, recover, inspect, or operate the Olympus multi-agent issue, repair, and repository-maintenance delivery environment for dm1681/Olympus. Use for Planner/Worker/Reviewer coordination, human or autonomous dispatch and merge controls, dirty-worktree recovery, one-off maintenance lanes, ownership-aware PR review, visual evidence and presentation, task organization, or selection of the next ready-for-agent issue.

  • Sm

    yansircc/skill-manager/skill/sm Skill

    0★ repo

    Manage the local SM skill registry, Producer-owned skill artifacts, Agent grants, immutable projections, and the Svelte dashboard. Use when an agent needs to inspect, add, generate, update, publish, authorize, build, apply, verify, or troubleshoot skills governed by ~/.sm for Codex, Claude, or Pi, or when the user asks to open or show the skill-management page so they can inspect or operate it themselves.

  • Worldcup linelock

    edycutjong/linelock/skills/worldcup-linelock Skill

    0★ repo

    Buy and audit World Cup betting edges from LineLock over Injective x402. Use when the user wants the next knockout pick (model probability vs market odds + a conviction staking ladder), wants to pay per-call in USDC without an API key, or wants to audit a tipster's settled record by closing-line value (CLV). Handles: checking USDC balance via the Injective MCP server, paying the 0.05 USDC x402 quote, validating the returned pick hash, and reading the free public ledger to compute trailing CLV before trusting a pick.

  • Fable legacy

    gz-ang/fable-legacy Skill

    0★

    Model orchestration doctrine authored by Claude Fable 5 — teaches any model to work like Fable (contract-first, whole-horizon planning, token frugality, evidence-based completion), routes tasks to the right Claude model (Opus 4.8, Sonnet 5, Haiku 4.5, Fable 5), defines how an orchestrator plans, dispatches parallel subagents, and reviews their work, and gates what comes back — how to judge a subagent's findings and its claims to have verified them. Use when choosing a model, spawning subagents, running parallel sessions, planning multi-agent work, when the user asks "which model should do this", when deciding whether to trust what a subagent reported back ("tests pass", "build green", "nothing found", "done"), or at the start of any large feature that will be split across agents.

  • Injective mcp servers

    ChiJian28/Touchline/.agents/skills/injective-mcp-servers Skill

    0★ repo

    Set up and run Injective MCP servers. Needed by multiple Injective skills which perform MCP tool calls.

  • Add skill

    Paldom/stealth-browser-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

  • Azpg ai in database

    lestermarch/postgres-ai-experts/skills/azpg-ai-in-database Skill

    no license0★ repo

    Call large language models and Azure AI services from inside Azure Database for PostgreSQL Flexible Server using the azure_ai extension — generate vector embeddings (azure_openai.create_embeddings), run LLM text generation (azure_ai.generate), and use Azure Cognitive Services for sentiment, language detection, key-phrase extraction, entity recognition, and summarization (azure_cognitive.*), all in SQL. Use this skill whenever the task involves in-database AI, calling Azure OpenAI or a Language service from SQL, embedding a column of text, enriching or summarizing rows with an LLM, configuring azure_ai.set_setting endpoints/keys, or bulk-generating embeddings for a table — even when the user just says "add embeddings to my products table" or "summarize these reviews in the database". For designing the vector index/RAG retrieval side hand off to azpg-pgvector-rag.

  • Manuscript review revision

    Jameslxr/manuscript-review-revision-skill/manuscript-review-revision Skill

    0★ repo

    Run a platform-independent, journal-aware, review-first workflow for scientific manuscripts, including target-journal confirmation or Top-5 journal recommendation, live official-author-guideline research, at least five independent reviewer agents, scientific and statistical review, claim-reference verification, reviewer-response drafting, revision, formal DOCX/PDF manuscript formatting, and fail-closed release auditing. Use in Codex, Claude Code, or another compatible Agent Skills host when the user asks to review, revise, polish, format, submit, retarget, recommend journals for, verify references in, or respond to reviewers about a biomedical or scientific manuscript. Always establish the exact target journal before full review; never revise or cosmetically polish before the independent review gate is complete and the user authorizes revision.

  • Add skill

    Paldom/promptimize/.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

  • Route github agent decisions

    Mimirs402/bountyverdict-mcp-skill/skills/route-github-agent-decisions Skill

    0★ repo

    Use for: why did a GitHub Actions run fail (root cause); retry a failed workflow; is a GitHub bounty open; which bounty to work on; audit AGENTS.md; will MCP tools/list changes break clients.

  • Jebi agent

    kyutomatte/jebi_agent/skill/jebi-agent Skill

    0★ repo

    Create a clean successor Codex chat for the current project by generating a concise handoff file and using Codex App Server thread/start. Use when the user invokes $jebi, $jebi-agent, JEBI AGENT, 제비, or asks in Korean to move/hand off/continue the current Codex work in a new chat: 제비로 옮겨주세요, 제비로 이사갈게요, 제비한테 넘겨줘, 제비로 인수인계해줘, 제비 출발, 새 채팅으로 제비 보내줘, JEBI로 새 채팅 만들어줘. Do not use for ordinary questions about swallows, images, filenames/code containing jebi, requests to explain JEBI AGENT, or when the user says not to move yet.

  • Shared obsidian brain

    aigeekops/shared-obsidian-brain Skill

    no license0★

    把任意 AI agent 经 MCP 接入 Obsidian vault,搭建本地优先的「第二大脑」 (Karpathy LLM Wiki 范式)。触发词:搭第二大脑 / second brain / LLM Wiki / 接入 Obsidian / 知识库。