agentsclimarketplace
Sign in

Shelves Context & AI engineering

Agents & orchestration

3,106 rows from 1,088 repositories

More than one agent, or one agent running long enough to need structure.

What Agents & orchestration skills agree on

969 skill files read, by 848 of the 1,300 authors on this shelf whose files we hold, 2026-09-06

The middle one of the 103 measured here is ~1.2k tokens long, counted with cl100k_base

Counted by distinct author, so one author publishing three of these counts once. Where a claim sits in fewer files than authors, that is said: a claim held by forty authors across three files is one file people copied, not forty people who agreed. Near-identical wordings are grouped and the other wordings are shown, so the grouping is yours to check.

What they tell the agent to do

  1. Dispatch one agent per independent problem domain56 of 848 in 42 filesalso worded as assign one agent per independent domain; Dispatch one Agent per phase
  2. Run full test suite after integration55 of 848 in 42 filesalso worded as run full test suite after agents return; Run full test suite
  3. Verify fixes do not conflict40 of 848 in 32 filesalso worded as Verify agent fixes do not conflict; Verify that agent fixes do not conflict
  4. Review each summary when agents return40 of 848 in 31 filesalso worded as Read each summary when agents return; Review each agent summary upon return
  5. Write a handoff document summarising the current conversation30 of 848 in 25 filesalso worded as Write a handoff document summarizing the current conversation; Write a handoff document summarizing the conversation
  6. Reference existing artifacts by path or URL26 of 848 in 24 filesalso worded as Reference other artifacts by path or URL; Reference artifacts by path
  7. Give each agent a specific scope19 of 848 in 10 filesalso worded as Provide specific scope to each agent; provide specific scope for each agent
  8. Give each agent a clear goal19 of 848 in 10 filesalso worded as Provide clear goals to each agent; provide clear goal for each agent
  9. Include a suggested skills section in the document18 of 848 in 16 filesalso worded as Include a suggested skills section in the summary; Include a suggested skills section
  10. Tailor the doc to the user arguments18 of 848 in 15 filesalso worded as Tailor the document to user arguments if passed; Tailor the document to any user arguments
  11. Issue all subagent dispatches in the same response17 of 848 in 11 filesalso worded as issue multiple subagent dispatches in the same response; Issue subagent dispatches in the same response
  12. Use git worktrees for isolation17 of 848 in 8 filesalso worded as Use git worktrees for isolated changes; Use git worktrees
  13. Save the document to the OS temporary directory17 of 848 in 16 filesalso worded as Save to the temporary directory of the user's OS; Save to the OS temp directory
  14. Remove the worktree after merging16 of 848also worded as Remove the worktree and branch.; Clean up worktrees after merging
  15. Write a handoff document15 of 848also worded as Write the handoff file; Write the handoff document to the memory path.

What they tell it not to do

  1. Do not use parallel agents when failures are related38 of 848 in 27 filesalso worded as Do not use parallel agents for related failures; Do not use parallel agents when failures are related
  2. Do not inherit session context or history35 of 848 in 27 filesalso worded as Do not let agents inherit session context; do not inherit session context
  3. Do not duplicate content already captured in other artifacts31 of 848 in 26 filesalso worded as Do not duplicate content captured in other artifacts; Do not duplicate other artifacts
  4. Do not just increase timeouts26 of 848 in 22 filesalso worded as Do not increase timeouts without finding the real issue; Do not just increase timeouts to fix timing issues
  5. Do not use when agents would interfere with each other18 of 848 in 12 filesalso worded as Do not use when agents would interfere with each other; do not use when agent interferes with another
  6. Do not parallelize tasks that depend on each other17 of 848 in 8 filesalso worded as Do not parallelize dependent tasks; Don't parallelize tasks that depend on each other's output
  7. Do not save to the current workspace14 of 848 in 13 filesalso worded as Do not save to the workspace; Do not save the document to the current workspace
  8. Do not dispatch parallel agents for related failures13 of 848 in 10 filesalso worded as do not use parallel dispatch if agents would interfere; Do not parallelize related failures
  9. Do not use parallel agents when there is shared state13 of 848 in 9 filesalso worded as Do not use when shared state exists; Do not use parallel agents when shared state exists
  10. Do not include sensitive information12 of 848 in 11 filesalso worded as Do not include sensitive info; Do not include sensitive data

What they expect to be installed

  1. git229 of 848
  2. bash67 of 848 in 62 files
  3. Task61 of 848 in 48 files
  4. Agent59 of 848
  5. gh57 of 848 in 53 files
  6. npm37 of 848 in 28 files
  7. Read36 of 848 in 30 files
  8. grep33 of 848
  9. python33 of 848
  10. python332 of 848

What they ask it to produce

  1. Handoff document79 of 848 in 74 filesalso worded as handoff summary; handoff note
  2. Summary of what you found and fixed52 of 848 in 38 filesalso worded as Summary of what was found and fixed; summary of findings and fixes
  3. Git worktree31 of 848 in 26 filesalso worded as isolated git worktree; worktree
  4. HANDOFF.md31 of 848also worded as session-handoff.md; handoff_active.md
  5. Pull request30 of 848 in 24 filesalso worded as Pull requests; Pull request summary
  6. report.md14 of 848also worded as findings.md; research.md
  7. Summary12 of 848also worded as SUMMARY.md; user summary
  8. CHANGELOG.md10 of 848 in 4 filesalso worded as change_log.md; docs/CHANGELOG.md

When Agents & orchestration authors say to reach for one

The situations these authors wrote into their own files, counted out of the same 848 authors, with the skills that name each one

How Agents & orchestration skills are built

2,779 skill directories by 980 authors, read from their repositories’ own file trees 2026-08-05

The middle bundle among those shipping files is 2 files, 16.7 KB beside SKILL.md

Counted by distinct author, same as above, so one author publishing forty template copies counts once. SKILL.md itself is not counted as a file, so a single-file skill is one where that file is the whole skill.

The shape

  1. SKILL.md is the whole skill424 of 980 authors, 1,458 of 2,779 skills
  2. files ship beside it556 of 980 authors, 1,321 of 2,779 skills
  3. executable scripts ship inside215 of 980 authors, 421 of 2,779 skills

The folders they converge on

  1. references/317 of 980 authors, 561 of 2,779 skills
  2. scripts/178 of 980 authors, 365 of 2,779 skills
  3. agents/151 of 980 authors, 241 of 2,779 skills
  4. assets/74 of 980 authors, 108 of 2,779 skills
  5. evals/53 of 980 authors, 91 of 2,779 skills
  6. templates/44 of 980 authors, 52 of 2,779 skills

  • Agent team design

    agentlas-ai/Agentlas-OS/skills/agent-team-design Skill

    1,154★ repo

    Agent OS: keep specialist agents in a hub, spin up a temporary orchestrator per task. Local-first, works with any model.

  • Choosing swarm patterns

    AgentWorkforce/relay/.agents/skills/choosing-swarm-patterns Skill

    793★ repo

    Real time communication for agents. Wake on message, channels, DMs and actions. Useful for orchestrating agents.

  • DirectorAgents 1

    momozi1996/awesome-ai-persona-skills/DirectorAgents-1 Skill

    555★ repo

    全网最全的、持续更新的、最火爆的 100+ 人格 蒸馏skills 合集｜ 多agent系统 ｜名人/导演/天涯大神/古籍/二次元/职场/情感全品类

  • Assistant orchestrator

    ZhixiangLuo/10xProductivity/.claude/skills/assistant-orchestrator Skill

    464★ repo

    Personal AI assistant for work inside corporate constraints, built on coding agents and the tools, sessions, and permissions you already have.

  • Ruv swarm

    ruvnet/ruv-FANN/io.github.ruvnet/ruv-swarm MCP server

    373★ repo

    Neural network swarm orchestration with WebAssembly acceleration and MCP integration

  • MCP Task Orchestrator

    jpicklyk/task-orchestrator/io.github.jpicklyk/task-orchestrator MCP server

    200★ repo

    Server-enforced workflow discipline for AI agents: work items, dependency graphs, quality gates

  • Ultracode

    PabloNAX/ultracode-skill/ultracode Skill

    174★ repo

    Codex-first Ultracode skill for dynamic coding workflows

  • Agent delegation

    kangarooking/system-prompt-skills/agent-delegation Skill

    162★ repo

    从 165 个顶级 AI 产品系统提示词中蒸馏出的 15 个可执行 Agent skill

  • Handoff codex

    dazuiba/handoff/cli/skills/handoff-codex Skill

    no license83★ repo

    Delegate tasks to DeepSeek right inside your Claude Code / Codex sessions.

  • Amq cli

    avivsinai/agent-message-queue/skills/amq-cli Skill

    81★ repo

    File-based message queue for local agent-to-agent communication (Maildir-style)

  • Devin handoff

    club-cog/devin-handoff/.agents/skills/devin-handoff Skill

    56★ repo

    Hand off tasks to Devin — a plugin/skill for Claude Code, Codex, Cursor, and any coding agent

  • Agent builder

    keysersoose/claude-agent-builder/agent-builder Skill

    41★ repo

    Build production-ready AI agents in Claude Code from natural language. Analyzes your problem, researches best practices, and generates subagents, skills, hooks, commands & MCP configs.

  • Handoff

    REMvisual/claude-handoff/skills/handoff Skill

    38★ repo

    Claude Code skill for session handoffs — survives context compaction, chain-links across sessions.

  • DirectorAgents

    momozi1996/DirectorAgents Skill

    38★

    "Multi-Agent Team — 让31位导演大师成为你的AI创作团队"

  • Gossipcat

    gossipcat-ai/gossipcat-ai/io.github.ataberk-xyz/gossipcat MCP server

    38★ repo

    Multi-agent orchestration for Claude Code — consensus review, adaptive dispatch, skill learning

  • Handoff skill

    ToolMonsters/handoff-skill Skill

    35★

    Claude skill: turn your current conversation into a complete handoff document so any LLM can pick up exactly where you left off

  • Proletariat cli

    chrismcdermut/proletariat/io.github.chrismcdermut/proletariat-cli MCP server

    35★ repo

    Agent orchestration platform - multi-agent spawning, tickets, boards, and workflows

  • Airflow and workflow orchestration

    vaquarkhan/data-engineering-agent-skills/skills/airflow-and-workflow-orchestration Skill

    29★ repo

    Production-grade Agent Skills for data engineering AI agents: 73 workflows, platform presets, safe backfill/replay, Kafka & Spark reliability, MCP observability, and VS Code/JetBrains installers.

  • Handoff

    arimet/handoff/skills/handoff Skill

    27★ repo

    Claude Code skill that quizzes you on the code Claude wrote — verify you understood it before shipping.

  • Loop maker

    jbrazy480/loop-maker/loop-maker Skill

    25★ repo

    🔁 Write loops, not prompts. Free skill that generates overnight autonomous AI runs for Claude Code, Codex & Cursor: master prompt, plan, verifier, one command. By AI Guy Official.

  • Agent graphs

    launchdarkly/ai-tooling/skills/agentcontrol/agent-graphs Skill

    no license25★ repo

    LaunchDarkly's official AI tooling

  • Senior fable

    AndyShaman/senior-fable/skills/senior-fable Skill

    23★ repo

    Tech-lead orchestration for Claude Code — the top-tier model (Fable) keeps architecture & decisions, cheap subagents (Sonnet/Opus) do the routine and the digging. Save tokens without losing quality.

  • Agentic delegation

    Raishin/vanguard-frontier-agentic/.claude/skills/agentic-delegation Skill

    20★ repo

    Curated marketplace of AI skills, agents, and rules for cloud, zero-trust, and compliance-aware engineering - works with Claude Code, Codex, Cursor, Copilot, and more.

  • Claude fable 5 spanking claude opus 4.8

    erhantezakar/claude-fable-5-spanking-claude-opus-4.8 Skill

    14★

    SKILL: Claude FABLE 5 spanking Claude OPUS 4.8. Orchestrator-worker cost model for Claude Code: the expensive model reads and judges, Opus 4.8 subagents do all the writing.

  • Build it

    DevOtts/build-it Skill

    14★

    Autonomous delivery orchestrator for Claude Code — hand it a goal and a numbered Definition of Done, it runs the whole job to completion

  • Multi agent researcher

    ahmedibrahim085/Claude-Multi-Agent-Research-System-Skill/.claude/skills/multi-agent-researcher Skill

    11★ repo

    A Multi-Agent Research System using Claude Code Skills. it is inspired by Anthropic claude-agent-sdk-demos

  • Adk expert

    jpantsjoha/ai-native-developer-experience/.agents/skills/adk-expert Skill

    11★ repo

    Team-wide AI harness adoption plugin, \w operating model, onboarding and delivery standards coherent human-agent outcomes from day one.

  • 01 super lab

    animaresearch/skills/01-super-lab Skill

    no license10★ repo

    Public AI skills: small, useful agent workflows you can try today.

  • Acceptance orchestrator

    sinhoneyy/master-skills/plugins/master-skills-ai-agent-builder/skills/acceptance-orchestrator Skill

    no license10★ repo

    Unified skill library for Claude, Codex, Cursor, Antigravity & AI agents — 2,658 skills across 15 domains

  • Genpark accio csuite manager

    alphaparkinc/genpark-accio-csuite-manager Skill

    no license9★

    GenPark Distilled Skill: accio-csuite-manager

  • Genpark nano banana video orchestrator

    alphaparkinc/genpark-nano-banana-video-orchestrator Skill

    no license9★

    GenPark Distilled Skill: nano-banana-video-orchestrator

  • Powabase

    powabase-ai/agent-skills/skills/powabase Skill

    9★ repo

    Agent Skills that help AI coding assistants build on Powabase — the AI Backend-as-a-Service (RAG, agents, orchestration, workflows) plus a Supabase-style BaaS layer.

  • Orchestrator

    hamza-ali-shahjahan/hamzaish/_archive/v0/agents/_orchestrator Skill

    8★ repo

    Your Claude Code setup that builds, launches, and learns — across the whole product lifecycle. Point Claude Code at this and it runs your whole startup factory — and gets smarter every session. Works with Cursor, Codex & Windsurf too.

  • Token lean

    hurttlocker/token-lean/skills/token-lean Skill

    8★ repo

    The decisions-only orchestration discipline for AI agent fleets. One markdown skill, any model — Claude Code, Codex, Cursor, anything that reads markdown.

  • Handoff revive

    sofumel/claude-handoff-revive/plugins/handoff-revive/skills/handoff-revive Skill

    6★ repo

    Resume Claude Code work after rate/usage/context limits without replaying the prior transcript. Auto-saves at 90%/95% usage. Plugin-installable, 10 languages.

  • Dynamic workflow patterns

    alex-feel/claude-code-artifacts-public/skills/library/dynamic-workflow-patterns Skill

    6★ repo

    Reusable Claude Code artifacts — skills, hooks, rules, environment configs, agents, and slash commands — ready to drop into any setup.

  • Graph engineering

    VineeTagarwaL-code/graph-engineering Skill

    6★

    Claude agent skill that plans multi-step work as a dependency graph. Subagent fan-out, layered fan-in, model tiering, verification retry loops, human approval gates. No framework, just instructions.

  • Code forge

    tercel/code-forge Skill

    no license5★

    Complete development workflow — from TDD-driven implementation plans to execution, debugging, code review, git worktree management, branch lifecycle, and parallel agent dispatch.

  • Swarm fan out

    ben-alkov/the_swarm/src/skills/swarm-fan-out Skill

    no license5★ repo

    Claude Code plugin for multi-pattern multi-agent orchestration, using CC's built in Agent Swarm as the engine — dispatches and coordinates specialist sub-agents in fan-out, pipeline, map-reduce, speculative, and swarm configurations.

  • Skills

    2389-research/building-multiagent-systems/skills Skill

    5★ repo

    Architecture patterns for multi-agent systems with orchestrators, sub-agents, and tool coordination

  • Catalyst orchestrator

    shalintripathi/saas-marketing-agents/plugins/saas-marketing/skills/catalyst-orchestrator Skill

    5★ repo

    Open-source AI marketing team for B2B SaaS - 59 specialist agents, 13 installable Claude Code skills, and the CATALYST orchestrator, with worked ABM/launch/demand-gen workflows and an AEO/GEO playbook. MIT.

  • Abm program orchestrator

    Stallin-Sanamandra/b2b-saas-marketing-skills/skills/abm-program-orchestrator Skill

    5★ repo

    B2B SaaS marketing skills for AI agents. Built for Claude, Cursor, and any SKILL.md-compatible agent.

  • Swarm debug

    AnmarHani/SwarmVault/skills/swarm-debug Skill

    5★ repo

    A shared knowledge vault + full software-engineering workflow for AI coding agents — run Claude Code and Codex in parallel with one memory, one plan.

  • Factory droid

    magnus919/hermes-droid/factory-droid Skill

    5★ repo

    Standalone Agent Skill for delegating coding work to Factory Droid

  • Agent teams

    christopherlouet/claude-base/.claude/skills/agent-teams Skill

    5★ repo

    Opinionated Claude Code foundation — Explore → TDD → Audit workflow, auto-detected stack presets (nextjs, fastapi, astro, ...), curl | bash install. MIT.

  • Mcp orchestrator

    rupinder2/mcp-orchestrator/io.github.rupinder2/mcp-orchestrator MCP server

    5★ repo

    Aggregates tools from multiple MCP servers with unified BM25/regex search and deferred loading.

  • Arch decision

    jsingh6/arch-decision/skills/arch-decision Skill

    no license4★ repo

    Open-source ADR orchestrator — explores your codebase, proposes approaches, writes Architecture Decision Records automatically

  • Autonomous loop

    BechsteinDigital/claude-shiploop/autonomous-loop Skill

    4★ repo

    Autonomous delivery skills for Claude Code — pitch an idea, get an MVP. CEO/PO/DEV/Reviewer agent team with contracts, gates, and evidence rules.

  • Handoff

    ananthanandanan/skills/skills/engineering/handoff Skill

    4★ repo

    Skills I use for my agentic engineering workflow.

  • Benchmark agents

    build-with-dhiraj/ai-workflow-framework-portability-kit/Plugins/vercel-marketplace-source/.claude/skills/benchmark-agents Skill

    4★ repo

    Portable, self-contained snapshot of a complete Claude Code setup — 36 specialist agents, 134 skills, plugins, MCP servers & host tooling. Clone, claude login, run one script, restore the whole orchestration stack in ~20 min.

  • Goodcode

    SpinaBuilds/goodcode Skill

    4★

    Host-agnostic exhaustive multi-agent orchestration as a portable Agent Skill — the /ultracode method on any coding agent that can spawn subagents.

  • Multi agent governance

    levelsofself/mcp-nervous-system/skills/multi-agent-governance Skill

    4★ repo

    The Nervous System - LLM Behavioral Enforcement Framework. Anthropic built the brain. This is the nervous system.

  • Cross model handoff

    ntaffzii/Skill-Agents/Skill.md/cross-model-handoff Skill

    4★ repo

    Personal AI-agent skills, workflows, prompts, and setup guides for local LLMs and MCP-powered assistants.

  • Handoff

    wassermanproductions/master-canvas-hermes-plugin/skills/handoff Skill

    4★ repo

    Hermes Agent plugin and bundled skill for Master Canvas AI video handoff packages.

  • Coding agents superpack

    frankxai/awesome-hermes-agent-skills/skills/coding-agents-superpack Skill

    no license4★ repo

    Awesome list of Hermes-compatible skills & plugins from across the web (+ small free open-core packs) | agents companion: frankxai/awesome-hermes-agents

  • Ab scripting feature dev

    mumez/pharo-agentic-browser/skills/ab-scripting-feature-dev Skill

    no license4★ repo

    Multiple AI agent session browser for Pharo Smalltalk

  • Epic orchestrator

    chancegraff/rules-for-claude/skills/epic-orchestrator Skill

    no license3★ repo

    Hooks, rules, skills, and instructions for use with Claude Code

  • Adversarial review

    AlexShchuka/neuro-matrix/skills/adversarial-review Skill

    3★ repo

    Claude Code plugin: runtime invariants + hooks (cycle-detector, approval gate, verification gate), four co-system agents (developer · analyzer · critic · epistemic-auditor), and a held-out evaluation harness. Anti-hallucination protocol grounded in game theory and the scientific method.

  • Huashu agent swarm en

    Biraj2004/huashu-skills-english/huashu-agent-swarm-en Skill

    3★ repo

    Huashu's Content Creation Skills Collection — 21 Practical Skills translated to English using Claude.

  • Agent creator

    coroboros/agent-skills/skills/agent-creator Skill

    3★ repo

    AI agent skills for Claude Code and compatible agents — workflow, coding, design, Claude Code meta, media, productivity, and writing

All 3,106 in the catalog