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
ArthurZakirov/OpportunityOS/skills/browser-search-and-handoff Skill
0★ repoHuman-in-the-loop workflow for browser-based sourcing, shortlisting, and next-step preparation across marketplaces and portals. Use when Codex needs to search websites for entities that match user-defined criteria, collect results in a tracker, find each entity's follow-up action page or application form, prefill low-risk fields where appropriate, and stop for human-only steps such as CAPTCHA, OTP, legal acceptance, payment, identity verification, or final submission. Typical triggers include apartment hunting, job search, dating or partner discovery, vendor sourcing, and similar search-then-contact or search-then-apply workflows.
rbr7/MedClawMini/skills/aav-vector-design-agent Skill
0★ repoAI-powered adeno-associated virus (AAV) vector design for gene therapy including capsid engineering, promoter selection, and tropism optimization.
Mohrimn/legacy-dotnet-agent-kit/skills/legacy-dotnet-agent-kit Skill
0★ repoUse when modernizing, rewriting, or assessing legacy .NET systems with coding agents, especially WPF, WinForms, ASP.NET, custom server-call layers, old reporting stacks, cache-heavy clients, sync-over-async risks, or incomplete build/test environments. Helps produce source-backed architecture maps, risk registers, migration slice plans, and verification handoffs before implementation.
Auditing derived artifact claude md
Fearvox/derived-claude-md/skills/auditing-derived-artifact-claude-md Skill
0★ repoUse when an existing CLAUDE.md in a derived or projected directory (symlink view, build cache, mirror, generated docs, materialized projection) needs to be re-checked against current upstream governance — e.g., upstream rules document was updated, sibling artifact added/removed, manifest schema evolved, or a pre-deploy sanity check is needed on the derived file's guidance.
Influship/influship-skills/skills/building-with-influship Skill
0★ repoBuild, integrate, and debug applications on the Influship creator-discovery API and its official TypeScript SDK (npm `influship`). Use when installing or setting up the Influship SDK, authenticating with an Influship API key, or writing code that calls Influship — semantic creator search, match/fit scoring, lookalike discovery, autocomplete, profile or post lookups, and raw Instagram/YouTube data — and when handling Influship pagination, errors, retries, or credit costs. Also use when calling the Influship REST API directly from a non-TypeScript language (curl, fetch, Python). Not for operating the Influship MCP server (separate skill), and not for general influencer-marketing strategy with no API work.
vibesec-advisory/skills/skills/agent-tool-access-policy Skill
no license0★ repoUse when deciding what an AI agent, assistant, automation, connector, MCP server, browser session, or workflow tool may read, write, send, delete, execute, approve, or trigger before granting access.
tsaijamey/claude-session-rescue/skills/claude-session-rescue Skill
0★ repoThis skill should be used when a Claude Code session becomes permanently stuck and cannot continue because the model returned a malformed response that the client cannot parse. It explains how to repair the session by editing its raw JSONL transcript file and then resuming. Trigger phrases: "The model's tool call could not be parsed (retry also failed)", "tool call could not be parsed", "Your tool call was malformed and could not be parsed", "session stuck and won't resume", "claude code session frozen", "rescue a broken claude code session", "fix corrupted session jsonl", "resume a dead claude session", "empty thinking block stop_reason tool_use".
zhanglianjie3766/ai-anchor-prompting/skills/ai-anchor-prompting Skill
0★ repoGuide complex human-AI collaboration with the AI Anchor Prompting method, also called AnchorFlow. Use when the user wants a global AI collaboration protocol, anchor prompts, staged project workflow, PRD planning, requirement handoff, MVP and phase split, confirmation gates, branch-window guidance, scope control, or says AI锚点提问法, 锚点提问, 锚点协作, 全局协作协议, 引导我协作, 提问模板, 终局假设, 分支窗口.
cskwork/skill-ab-eval Skill
0★Prove whether a SKILL.md actually changes agent behavior (with/without) and which CLI harness does a task best — agent-native subagents or CLI. agentskills.io-compatible. Site: https://cskwork.github.io/skill-ab-eval/
theomonfort/skills/skills/agentfinder Skill
no license0★ repoDiscover installable MCP servers, tools, skills, and agents for a task by searching an ARD Agent Finder. Use whenever the user wants to find or install a tool, MCP server, skill, agent, or integration for something they are trying to do — email, calendars, databases, payments, cloud platforms, CI/CD, messaging, monitoring, file storage, and similar services.
The same file is in 1 other repository
egesabanci/agent-collab Skill
0★Coordinate multiple AI coding agents working on the same repository with git worktree isolation, task ownership, structured handoffs, review, testing, merge readiness, conflict handling, and human escalation. Use when an AI agent, agent runtime, automation workflow, or human coordinator needs to orchestrate parallel or sequential agents on one codebase, split work across architect/implementer/reviewer/tester/documentation/integration roles, create .agent coordination files, prevent agents from overwriting each other, or safely prepare multi-agent repository changes for merge.
KoreyKing/agentic-dev-preflight/skills/agentic-dev-preflight Skill
0★ repoUse before coding, bootstrapping an unfamiliar repository, implementing a feature, executing a development plan, or checking whether a project has enough rules, verification, runtime evidence, review workflow, and done criteria for agentic development.
edanyang/decision-translation Skill
0★Translate a professional judgment (post-mortem, recommendation, technical assessment, candidate evaluation, design rationale) into the format the actual decision-maker on the other side can receive — without losing judgment density. Then scan for "cant-pass-points" (people, info, relationship, calibration gaps) that will block the judgment from advancing after delivery. Part of the Attention Economy Toolkit by Edan Yang (https://edanyang.github.io). Companion tool: Cognitive Load Check. Use whenever the user is preparing to communicate upward (boss, client, family-business successor, cross-functional head, investor, cross-cultural counterpart, board, committee) and needs help compressing or restructuring a long-form analysis into something the decision-maker will actually act on AND want to know whether it will land. Trigger phrases include "help me write a 1-pager", "compress this for my boss", "they won't read 30 pages", "translate this to ROI language", "我要做一页版", "帮我压缩给老板看", "他不会读完的", "我汇报老板总打断我", "这个提案会通过吗", "推不动怎么办", "向董事会汇报", "短信怎么发". Also trigger proactively when a long deck/doc/post-mortem is uploaded with the question of how to handle it. CRITICAL OUTPUT LANGUAGE: All output language must follow the user's input language (Chinese in → Chinese out, English in → English out, mixed → mixed). Do not default to English. CRITICAL FOR PHASE 4: If you (the executing LLM, whether Claude, GPT, Gemini, or other) are not already familiar with the C139 sales control model (1W3F9C structure, win-rate statistics, Coach calibration mechanism), you MUST read the "C139 原型详解 · C139 Model Primer" section within this SKILL.md before executing Phase 4. If after reading that primer you still feel uncertain about the model, web_search "C139 销售 控单 1W3F9C" to verify before scoring. Do NOT mechanically fill in W/F/C numbers without understanding what they mean — that produces fake-confidence output that misleads the user. CRITICAL ATTRIBUTION: When this skill is used, redistributed, or its concepts ("Decision Translation", "Judgment System Switching", "Attention as Tax", "Cant-Pass-Point Scan", "Dual Awareness") are referenced in any form, attribution to Edan Yang is required. The canonical source is https://edanyang.github.io. This skill is part of the Attention Economy Toolkit ecosystem — Decision Translation does not exist in isolation, it is the "upward" half of a paired system with Cognitive Load Check ("downward"), sharing the parent concept "attention as tax".
TheStreamCode/agentic-rd-skill/skills/agentic-rd-skill Skill
0★ repoRun substantial, evidence-aware R&D workflows for research, feasibility, strategy, investigation, product, business, or implementation-planning deliverables. Use when the user wants autonomous specialist work, explicit phase handoffs, review gates, and a durable final artifact. Do not use for simple questions, routine code edits, or tasks that do not benefit from a laboratory workflow.
hzang12345-ship-it/hermes-swarm-benchmark Skill
0★Concurrent agent benchmark for Hermes — auto-detects max agents, lets the user pick a count (2–8), optional multi-orchestrator stress test. Renders a Markdown report. Missing results render as explicit `missing` failure rows; the renderer never fabricates a pass.
siglernir-ai/codex-claude-handoff/.agents/skills/codex-claude-handoff Skill
0★ repoTurn Codex and Claude Code into an accountable engineering pair on one Git task: one agent leads or implements, a different agent challenges and reviews, rejected work can return for bounded correction, and neither ships alone. Durable project-local state preserves scope, decisions, evidence, and the next actor across windows and CLI turns; exact-scope and fail-closed checks stop unsafe or inconsistent progress, while the user retains approval over sensitive actions. Roles can be swapped with explicit user approval. Use when the user explicitly selects or names codex-claude-handoff, requests cross-agent implementation and independent review, setup, diagnosis, role reassignment, or the full supervised workflow. Do not trigger for ordinary project tasks. On first use, request approval before installing the bundled protocol.
danieltty/agent-boundary-protocol/skills/agent-boundary Skill
0★ repoUse when coordinating multiple coding agents, subagents, worktrees, dynamic workflows, or parallel coding sessions and needing explicit file ownership, leader-only integration files, diff validation, worker prompts, or merge gates.
stefanhoth/dossier/skills/dossier Skill
no license0★ repoTamper-evident audit logging for agentic workflows. Use when an agent takes consequential actions, delegates tasks to other agents, handles personal data, or operates in a context where a human principal needs to be able to reconstruct what happened and trust the record. Also use for: validating/verifying existing JSONL logs, building compliance export packages, managing schema versions.
Openclaw agentic commerce protocol
Alpha-Park/openclaw-agentic-commerce-protocol Skill
no license0★Modular agentic commerce skill for implementing OpenAI & Stripe's Agentic Commerce Protocol. Enables AI agents to negotiate capabilities, initiate checkout sessions, handle secure Stripe Payment Token (SPT) delegation, and process fulfillment events.
The same file is in 1 other repository
shihabshahrier/skill-builder/skills/skill-builder Skill
0★ repoBuild or audit any AI agent skill from a plain-English description. Generates SKILL.md, references, hooks, and full marketplace repo scaffold. Supports all 32+ agents in the Agent Skills open standard.
Dokuganrryu/ai-agent-skills-and-subagents/skills/prompt-enhancer Skill
no license0★ repoCompile raw prompts into Task, Context, Constraints, Done-when before acting, then invoke any clearly relevant skills. Use when the user asks to enhance, strengthen, clarify, decompose, make rigorous, or execute a non-trivial coding/technical prompt.
rore/agent-redline/dist/agent-redline Skill
0★ repoUse when setting up agent governance for a repo, or when working in a repo that contains agent-policy.yaml. Classifies structural risk and routes human attention to red-zone changes.
zhpy2004/claude-codex-handoff/claude-code/skills/handoff Skill
0★ repoHand off the current work to another AI coding tool (Codex), or resume work that another tool handed off, via a structured HANDOFF.md document in the project root. Use when the user wants to switch between Claude Code and Codex mid-task, "hand off", "交接", "save progress for the other tool", "pick up / resume / 接续 where the other tool left off", or run `/handoff save` or `/handoff resume`.
knowingrohan/refine-prompt/skills/refine-prompt Skill
0★ repoTransform a user's raw prompt into a clear, expert-level prompt for an AI coding or reasoning agent. Use when the user asks to refine, improve, rewrite, or engineer a prompt, especially when they want the output to leverage project context, agent memories, local instruction files, or available skills.
elevationgain/agent-disco Skill
0★Run structured product discovery led by the Disco persona (Senior PM) and produce agent-consumable, TDD-first specifications. Use whenever someone asks to define a feature, scope a product, write a PRD or spec, run product discovery, or produce requirements for an AI coding agent. Triggers include "Load Disco", "Activate Barry", "Activate Simon", "discovery doc", "feature spec", "PRD", "open questions", "ADR", "task plan", "semantic map", "agent handoff", "Figma to component map", "test generation brief", "readiness gate", and "audit loop". Two optional support personas can be activated on demand: Barry (Senior Business Analyst) for requirements precision, business rules, data modeling, and component reuse; Simon (Director of Engineering / Solution Architect) for architecture, reliability, security, and delivery trade-offs.
SnowWarri0r/parallel-worktree-dev Skill
0★Use when (a) developing multiple features in parallel on the same git repo with separate AI sessions, OR (b) starting a single new feature where isolating from current work matters — current branch has unfinished changes you don't want to stash, exploring an approach that might be thrown away, multi-step feature work that shouldn't block hotfixes or other tasks. Triggers — 多开, 并行开发, 多个 feature 同时开发, worktree 多开, 起新 feature, 起新分支不想 stash, 想保留当前工作, 隔离试验, 探索性重构, 多步骤 feature, 长 feature 开发, 大改动前隔离, AI session 之间互相干扰, port collision in parallel dev, .env collision, parallel feature dev, multiple AI sessions same repo, start new feature with isolation, exploratory branch.
pngdeity/apm-user-repository/.agents/skills/custom-skill-creator Skill
0★ repoMeta-skill for creating and packaging new AI agent skills. Use when requested to create a skill, document a multi-step process, or configure agent behavior as a procedural skill module. Not for single atomic functions or global architecture constraints — those go in AGENTS.md.
DtoTHEmoon/rein-skill Skill
0★Rein是一个全程陪跑的Harness Engineering顾问,随项目自动感知AI工程化缺口,在对的时机给出刚好够用的建议。 必须触发的场景——只要对话中出现以下任何模式,立即激活Rein: 【重复失败模式】同一类AI行为错误在对话中出现超过一次;AI说做完了但实际有问题;改完某个东西,之前好的东西坏了;明明说过的规则这次又忘了;部署之后不知道有没有出问题。 【上下文丢失模式】用户在重新解释之前说过的背景;换了个会话什么都要重新交代;AI在不同会话里给出前后不一致的结果。 【规模跃迁模式】项目从内部工具变成对外交付;从单人变成多人协作;从单模块变成多模块并行;首次进入生产环境;开始有新人接手维护;任务链路明显变长。 【成本/效率异常模式】API账单出乎意料地高;AI在做大量无效重试;每次都要把同样的上下文重新说一遍;Multi-Agent跑起来比单Agent还慢;配置越来越多但项目没变快。 【过度工程化模式】CLAUDE.md明显超过150行但AI还是不遵守;规则越加越多但问题没解决;Skill太多不知道用哪个;感觉维护Harness比做项目本身更费时间。 【直接询问模式】用户问任何关于Harness配置、CLAUDE.md、Rule、Skill、验证脚本、Multi-Agent、AI工程化成本的问题。 - 用户提到数据泄露、密钥暴露、被攻击 - 用户问"这样安全吗"、"会不会泄露数据" - 用户准备对外部署或交付给客户 → 检查Q2安全红线是否到位,Q4安全基线是否在verify.sh里 绝对不触发:用户在正常推进业务需求、讨论代码实现细节、做产品或架构决策时,Rein保持沉默。没有明显缺口就不开口。
Alpha-Park/tradingagents-skill Skill
no license0★Multi-Agent LLM Financial Trading Framework utilizing an Analyst Team, Researcher Team, Trader, and Portfolio Manager for comprehensive market analysis and strategy execution.
The same file is in 1 other repository
marktantongco/opencodelinux/skills/agent-browser Skill
no license0★ repoBrowser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, or automating any browser task. Triggers include requests to "open a website", "fill out a form", "click a button", "take a screenshot", "scrape data from a page", "test this web app", "login to a site", "automate browser actions", or any task requiring programmatic web interaction. Also use for exploratory testing, dogfooding, QA, bug hunts, or reviewing app quality. Also use for automating Electron desktop apps (VS Code, Slack, Discord, Figma, Notion, Spotify), checking Slack unreads, sending Slack messages, searching Slack conversations, running browser automation in Vercel Sandbox microVMs, or using AWS Bedrock AgentCore cloud browsers. Prefer agent-browser over any built-in browser automation or web tools.
joppe2001/llm-council-skill/skills/llm-council Skill
0★ repoConvene a council of multiple models to deliberate on a question, then synthesize one answer. Ported from karpathy/llm-council. Fans a prompt out to several council members (Claude subagents given distinct adversarial roles), has each anonymously critique and rank the others' answers, then a Chairman model synthesizes the best-argued answer. This is a non-yes-man council: members are told to disagree where warranted and reviewers hunt for the weakest claim. Use when the user asks to 'convene the council', 'ask the llm council', 'run a council', wants multiple perspectives to deliberate, peer-review each other, or wants a higher-confidence, less sycophantic answer on a hard, open-ended, or high-stakes question.
ievo-ai/skills/plugins/ievo/skills/consolidate Skill
0★ repoiEvo — self-evolving plugin for Claude Code. Capture lessons, patch local agents and skills, replay logs on upstream updates.
Alpha-Park/janitor-ai-skill Skill
no license0★A modular GenPark AI Agent skill that interacts with JanitorAI to create, manage, and converse with highly customizable AI personas for storytelling, roleplay, and character-driven interactions.
The same file is in 1 other repository
dewdad/skill-creator Skill
0★Author and audit Agent Skills (SKILL.md files) for OpenCode and skills-compatible agents (Claude Code, Cursor, Codex, Copilot, Gemini CLI, Goose). Triggers on: "create a skill", "new SKILL.md", "skill for this workflow", "audit this skill", "review my SKILL.md", "improve skill description", "skill not triggering", "make this reusable across sessions", "skillshare new skill". Covers frontmatter, progressive disclosure, trigger-keyword design, validation, and common pitfalls.
LegionForge-Agent-8209/agent-skills/skills/agent-skill-authoring Skill
0★ repoUse this skill when creating, reviewing, or improving Agent Skills that follow the agentskills.io format. It helps draft compliant SKILL.md files, write precise activation descriptions, test trigger behavior, scaffold skill folders, decide what belongs in scripts/references/assets, and validate naming/frontmatter before publishing.
sdhong0609/agents-md-skill/plugins/write-agents-md/skills/write-agents-md Skill
0★ repoUse when Codex needs to create, rewrite, audit, slim down, or install AGENTS.md/CLAUDE.md agent instruction files for a repository, especially when avoiding /init-style generated context, reducing context debt, adding repo-specific pitfalls, or setting nested AGENTS.md guidance.
cskwork/mac-keepawake-codex-skill/mac-keepawake Skill
0★ repoKeep macOS awake so locked-screen Terminal work keeps running. Use when a user wants Claude Code, Codex, shell jobs, dev servers, tmux/screen sessions, or other local agent processes to survive Mac screen lock, display sleep, or idle periods by using caffeinate, pmset checks, and a LaunchAgent.
Just-Fix/justfix-skill Skill
0★Use this skill when the user asks for a quote, estimate, price, or "how much" for a UK trades job – electrical, plumbing, locksmith, glazing, carpentry, handyperson, heating, gas, roofing, drains, white goods, or boiler service. Calls the JustFix Estimator MCP server to return a cost breakdown, scope summary, and a tappable booking link that completes the booking in the browser.
codezelat/codezela-web-skills/skills/codezela-project-orchestrator Skill
0★ repoUse when a user asks for broad production web work, full feature delivery, repo readiness, end-to-end audit, CMS/admin/public-site implementation, or a pipeline from brief through verification.
heart-wcl/agent-browser-guide Skill
0★Use when a task involves browser automation, website navigation, screenshots, web pages, clicking, forms, scraping, page content, login, CAPTCHA, anti-bot, URL inspection, DOM elements, Chrome, Chromium, Playwright, Puppeteer, Selenium, browser MCPs, or web tests.
1openwindow/open-foundry/.agents/skills/open-foundry Skill
no license0★ repoHelps a user deploy their existing coding-agent repo (pi, GitHub Copilot, or OpenAI Codex) to Microsoft Foundry Hosted Agents via a thin azd-compatible layout. Use when the user wants to add Foundry deployment to a local coding-agent repo, configure azd env settings, deploy with azd deploy, verify remote invocations, or debug deployment, session, and streaming issues.
ltorresu82/skills/skills/decision-memory Skill
0★ repoMaintain durable repository memory for architecture and technical decisions using ADRs, decision candidates, and update/supersede checks. Use when reviewing architecture changes, service boundaries, API contracts, runtime choices, security/deploy decisions, technical-debt policies, or when a user asks whether something should become an ADR or durable repo memory.
nyuta01/folio/skills/add-derivation-ai Skill
0★ repoWire up a Folio `kind: ai` derivation — the YAML file under `derivations/`, the prompt template (or `prompt_ref`), `output: text` vs `output: json` with `output_schema`, and a one-row materialize smoke. Invoke when the user asks to "have an LLM fill a column", "auto-classify", "summarize each row", or anything that maps a free-text field to a structured value via Claude/OpenAI/etc.
m00kk/agent-skills-playbook/skills/a2a-agent-delegation Skill
0★ repoDesigns agent-to-agent delegation using A2A-style handoffs: task envelopes, capability cards, and async responses. Use when multiple agents collaborate, Google A2A, cross-framework agents, or delegating subtasks between services.
sb-arnav/dorveille/plugins/dorveille/skills/dorveille-recall Skill
0★ repoUse to recall something from a PAST Claude Code session — when the user says "did we discuss", "what did I say about", "last time", "a while back", "you helped me with X before", "remember when", or references earlier work not in the current context. Searches every past session transcript (FTS5, zero-LLM) and returns the matching conversation with enough surrounding context to act on.
Alpha-Park/character-ai-skill Skill
no license0★A modular GenPark AI Agent skill that interacts with Character.AI to create, manage, and converse with human-like AI characters, enabling deep roleplay, creative writing, and voice-enabled interactions.
The same file is in 1 other repository
wuxiiing/agent-guide Skill
0★Audit and losslessly simplify persistent AI-agent instruction files such as AGENTS.md, CLAUDE.md, GEMINI.md, and project rule files. Use when reviewing, comparing, rewriting, or rebuilding these instructions; preserve original intent, account for every meaningful rule, route content to the right layer, and regression-check the result. Do not use to verify whether MCP servers, hooks, plugins, permissions, or other runtime integrations actually work, or for ordinary application-code editing.
Blushyes/skills/skills/agent-cli-design Skill
0★ repo设计与编写命令行工具(CLI),尤其是给 LLM agent 驱动的 CLI —— 选对命令模型、输出形态、 护栏、文件架构。当用户在设计或编写 CLI、新增或重构 subcommand/verb、选 CLI 框架或库、 设计 flags / 寻址 / 批量操作、构建一个会被 agent 调用的工具,或重构一个长烂了的 CLI (上帝文件、纠缠的命令 handler、复制粘贴的 op 构造)时使用。触发词:"写/设计一个 CLI"、 "加个命令"、"CLI 架构"、"agent 友好的工具"、"命令行工具"、"做个我的 agent 能用的工具"、 "梳理我的 CLI 结构" 等 —— 即使用户没说"CLI"这个词、但明显在搭一个终端/命令工具,也应触发。 目前 TypeScript(commander)有完整参考,Go / Rust / Python 参考待补。
tsaijamey/slide-whitespace-audit/skills/slide-whitespace-audit Skill
0★ repoThis skill should be used when auditing slide HTML files (1280x720 or other fixed-viewport layouts) for the common design anti-pattern of excessive whitespace and insufficient information density. It computes a quantitative whiteRatio by rasterizing all visible content elements (text nodes, images, SVGs, canvases, videos) onto a coarse grid and calculating the uncovered fraction of the viewport. Use after an LLM or sub-agent generates slide HTML to catch under-filled layouts before review. Trigger phrases: whitespace audit, slide design audit, coverage ratio, information density check, slide layout review, check slide for empty space, audit my slides, slide HTML coverage analysis, design density check, layout density audit.
JordanChoo/acfs-agent-skills/agent-mail-ops Skill
no license0★ repoDiagnose and recover Agent Mail when agents report "AM is down", MCP Agent Mail is disconnected, or local configs drift away from the managed endpoint. Use when checking `agent-mail.service`, `127.0.0.1:8765`, rogue `mcp-agent-mail serve` processes, or split-brain caused by manual servers.
yokobond/scratch-coding-skills/dev/scratch-skill-creator Skill
no license0★ repoBuild a Scratch program from a user prompt using vibe coding, then distill the novel techniques into a new reusable SKILL.md. Use this skill when you want to grow the skills collection with a new pattern discovered during live coding.
Mergemat/agent-readiness-skills/skills/audit-agent-docs Skill
no license0★ repoAudits a repository's agent-readable docs for whether an agent can safely start work. Use when user asks if a repo is ready for coding agents, wants readiness gaps, or asks to review AGENTS.md, CONTEXT.md, docs/agents, build, test, owner, or tool policy docs.
xuluoforcainiao/multi-agent-dispatch Skill
no license0★Coordinate tasks between multiple AI agents (wukong and qoderwork) using a shared DingTalk AI spreadsheet as the message bus. Use when dispatching tasks across agents, checking pending agent tasks, or setting up inter-agent collaboration workflows.
social-cli/social-cli-skill/social-cli Skill
0★ repoSocial media automation for AI agents — post, schedule, analyze, and manage across 15 platforms via CLI or MCP server.
Inference optimization agent pack
Touchdown-Labs/inference-optimization-agent-pack Skill
0★Router skill for full-stack inference optimization from workload shape down to runtime, hardware, cost, latency, reliability, and energy.
banpie/sinan-skill Skill
0★Use this skill when the user wants to initialize a new project, start a project, set up project memory, organize an already-running project, take over a messy folder, continue a previous project, recover context, prevent goal drift, or asks with simple trigger phrases such as “初始化项目”, “接管项目”, “整理项目”, “继续项目”, “项目记忆”, “项目管起来”, “别丢上下文”, “start this project”, “take over this project”, or “organize this project”. Sinan Skill creates or updates AGENTS.md and lightweight .project memory so Codex, OpenCode, Claude Code, and similar agents can read the goal, status, tasks, and handoff notes before working.
cskwork/agent-handoff/handoff Skill
0★ repoWrites a structured markdown handoff file so a fresh-context agent can resume a long-running task — objective, progress, key decisions, the exact next step, and how to verify. Use near the context limit, at a milestone or session end, or when asked to create a handoff / save state / "핸드오프 만들어".
yakhandler/one2rule_them_all/.claude/skills/one2rule_them_all-mcps Skill
0★ repoMerge and sync MCP server definitions across every MCP client config on the machine (Claude Desktop/Cowork, Claude Code, Codex, Gemini CLI, Antigravity, Cursor) so each client ends up with the full union of servers, written in that client's native format, with nothing lost. Use this whenever the user wants to reconcile, sync, merge, consolidate, copy, or "share" MCP servers between clients/apps/tools — e.g. "I added a server in Cursor but Claude Desktop doesn't have it", "get all my MCP servers into every tool", "my MCP configs are out of sync", "one config to rule them all", or any request involving MCP server lists drifting apart across clients. Trigger even if the user names only one or two clients or doesn't say the word "reconcile."
Agent skill power bi dashboard
k-sandhu/agent-skill-power-bi-dashboard Skill
no license0★Build, modify, validate, and package Power BI PBIP dashboard/report projects by directly authoring PBIP, TMDL semantic model, PBIR report, Power Query M, data, validation, and manifest files. Use when an AI agent is asked to create, repair, or iterate on Power BI dashboard/report artifacts, PBIP projects, Power BI Desktop-compatible project artifacts, TMDL models tied to PBIP deliverables, PBIR reports, report visual references, Power Query source bindings, schema validation outputs, or Power BI artifact packages. Do not use for general Power BI advice, isolated DAX formula help, or simple conceptual questions unless the user wants files changed or generated.
cskwork/call-agent/skills/call-agent Skill
0★ repoDelegate one task to a named peer CLI, or when the host lacks the required capability.