New: 98,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

Add Coordinator

by FeiZhuLulu in FeiZhuLulu/Agent-Bridge

Verify that a new host can act as an Agent Bridge coordinator over stdio MCP.

MITAuto-check passedAgent Workflows

Install Add Coordinator

skills CLI
$ npx skills add FeiZhuLulu/Agent-Bridge --skill add-coordinator -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install FeiZhuLulu/Agent-Bridge add-coordinator --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/FeiZhuLulu/Agent-Bridge.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/add-coordinator .claude/skills/add-coordinator && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
add-coordinator
GitHub stars
110
Token cost
~1.9k tokens
SKILL.md length
954 words
Files
1
Skills in repo
3
Repo updated
First seen
Licence
MIT

At a glance

Verify that a new host can act as an Agent Bridge coordinator over stdio MCP.

  • Works in 10 steps: stdio MCP config shape, scopes, and… → Whether MCP server instructions reach… → AGENTS.md / rules loading and size… → …
  • Adding a coordinator host
  • SKILL.md covers Before writing anything, Current hosts (verified…, Implementation checklist… and Definition of done, plus 1 more section
  • Calls uv and git

What it does

Add Coordinator is an agent skill from FeiZhuLulu/Agent-Bridge. Verify that a new host can act as an Agent Bridge coordinator over stdio MCP. Use when adding a coordinator host, checking MCP/skill/timeout/permission setup for Codex, Cursor, Kimi Code, ZCode, Grok Build, or Claude Code, or when a host can also run as a worker and nested dispatch must stay disabled.

Its SKILL.md is about 1.9k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Agent Workflows, covering MCP servers. It works with Model Context Protocol and Kimi. The repository describes itself as: Agent Bridge is a connector that links various local agents, enabling users to coordinate multiple other agents simply by interacting with a single powerful agent. It currently… The licence is MIT.

When your agent uses it

  • Adding a coordinator host
  • Checking MCP/skill/timeout/permission setup for Codex
  • A host can also run as a worker and nested dispatch must stay disabled

Example prompts

  • “/add-coordinator”

Workflow steps

10 steps, taken from the first numbered list in SKILL.md.

  1. stdio MCP config shape, scopes, and precedence. User file vs project file. Whether a native config and a .agents/mcp.json fallback merge…
  2. Whether MCP server instructions reach the model. Handshake text is one channel; if the host hides it, the skill and ORCHESTRATION.md must…
  3. AGENTS.md / rules loading and size limits. Path, walk order, concatenation, and any byte/token cap.
  4. Skill directories. Native user dir, project dir, generic ~/.agents/skills, plugin/marketplace paths, and whether discovery is automatic.
  5. MCP startup timeout and tool-call timeout, with units. Recommend a value that leaves room for wait_task default 180 s (Bridge recommends…
  6. MCP tool permission / approval. How to pre-allow only Agent Bridge tools without turning the whole host into always-approve. Note which…
  7. Environment inheritance and executable PATH. Does the host clear the MCP child env (Codex does)? Can it resolve a PowerShell function, or…
  8. Can this product also be a Worker? If yes, a Worker that inherits the user's Agent Bridge MCP config will start a nested Bridge. Prove the…
  9. Real closed loop on the live host: list_agents → dispatch_task → wait_task → get_result → same session_id continue → end_session. Verify…
  10. Separate automation from human steps. Unit tests and a project-level config fixture are automation. Opening a GUI host and sending one…

What it can do on your machine

Read from SKILL.md and the folder at commit 5560e06. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • uv
    • git

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md. Its commands use uv and git, which can reach the network depending on how they are called.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Add Coordinator loads about 1.9k tokens when it runs. Until then it costs about 80 tokens; SKILL.md has 954 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~80
When it runs · the whole SKILL.md, loaded when a task matches
~1.9k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from FeiZhuLulu/Agent-Bridge at commit 5560e06, republished under its MIT licence (© FeiZhuLulu). 954 words, ~1,902 tokens.

Download SKILL.mdSave it as .claude/skills/add-coordinator/SKILL.md (or your agent's skills folder).
name
add-coordinator
description
Verify that a new host can act as an Agent Bridge coordinator over stdio MCP. Use when adding a coordinator host, checking MCP/skill/timeout/permission setup for Codex, Cursor, Kimi Code, ZCode, Grok Build, or Claude Code, or when a host can also run as a worker and nested dispatch must stay disabled.

Adding a coordinator host to Agent Bridge

This is not add-worker.

  • add-worker connects a new execution CLI to Registry / adapters / agents.toml.
  • add-coordinator proves a new host can drive the existing workers through the same MCP tools.

Do not add an adapter, a worker block, or a new dispatch protocol for a coordinator. Do not use GUI automation. Do not treat "the config file was written" as an end-to-end pass.

Before writing anything

Answer every item below from the host's current official docs and, when possible, its source. Record units (seconds vs milliseconds) and scope (user vs project).

  1. stdio MCP config shape, scopes, and precedence. User file vs project file. Whether a native config and a .agents/mcp.json fallback merge or replace each other. Whether the settings UI paste format differs from the on-disk file.
  2. Whether MCP server instructions reach the model. Handshake text is one channel; if the host hides it, the skill and ORCHESTRATION.md must still be discoverable.
  3. AGENTS.md / rules loading and size limits. Path, walk order, concatenation, and any byte/token cap.
  4. Skill directories. Native user dir, project dir, generic ~/.agents/skills, plugin/marketplace paths, and whether discovery is automatic.
  5. MCP startup timeout and tool-call timeout, with units. Recommend a value that leaves room for wait_task default 180 s (Bridge recommends 600 s / 600000 ms). Document the host default and the short-poll fallback when the default is too short or unknown.
  6. MCP tool permission / approval. How to pre-allow only Agent Bridge tools without turning the whole host into always-approve. Note which keys are user-scoped and must not go in a project example.
  7. Environment inheritance and executable PATH. Does the host clear the MCP child env (Codex does)? Can it resolve a PowerShell function, or must command be a real executable absolute path?
  8. Can this product also be a Worker? If yes, a Worker that inherits the user's Agent Bridge MCP config will start a nested Bridge. Prove the server-side gate: nested list_agents shows runtime_context=worker and dispatch_enabled=false; nested dispatch_task (even user_requested=true) returns nested dispatch is disabled.
  9. Real closed loop on the live host: list_agents → dispatch_task → wait_task → get_result → same session_id continue → end_session. Verify the file on disk yourself. A timeout is not failure.
  10. Separate automation from human steps. Unit tests and a project-level config fixture are automation. Opening a GUI host and sending one prompt is human. Never call the latter "E2E passed".

Current hosts (verified against their public docs)

HostNative MCP fileSkill discoveryTool timeout fieldNotes
Codex~/.codex/config.toml [mcp_servers.*]~/.codex/skillstool_timeout_sec (seconds)Clears MCP child env; list env_vars.
Cursor~/.cursor/mcp.json or <repo>/.cursor/mcp.json~/.cursor/skills~45–60 s, not configurableShort-poll wait_task.
Kimi Code~/.kimi-code/mcp.json~/.kimi-code/skillstoolTimeoutMs (ms)Default 60 s without override.
ZCodeuser ~/.zcode/cli/config.json (mcp.servers); project <repo>/.zcode/config.json~/.zcode/skillstimeoutMs (ms)UI "完整配置" accepts {name:{...}} or {mcpServers:{...}}. That is not the native file shape. Same-scope .zcode MCP skips .agents/mcp.json entirely.
Grok Builduser ~/.grok/config.toml; project <repo>/.grok/config.toml~/.grok/skills, ./.grok/skills, and ~/.agents/skillsstartup_timeout_sec / tool_timeout_sec (seconds; tool default 6000)Also a Worker. MCP stdio inherits the process env (no env_clear). Official permission form is [[permission.rules]] with tool = "mcp" and {server}__{tool} names. [ui] permission_mode is user-scoped only.
Claude Codeuser ~/.claude.json top-level mcpServers; project <repo>/.mcp.json; local scope nests under the project path in ~/.claude.json~/.claude/skills (does not read ~/.agents/skills)per-server timeout (ms); MCP_TOOL_TIMEOUT env; CLI default ~28 hAlso a Worker (claude-agent-acp, not product claude). Do not put mcpServers in settings.json. Pre-allow with permissions.allow = ["mcp__agent-bridge__*"]. Native rules file is CLAUDE.md.

Bridge already installs the coordinator skill into .agents, .cursor, .codex, .kimi-code, .zcode, and .claude. Grok reads ~/.agents/skills; do not add a .grok copy unless a future host stops reading .agents.

Show full SKILL.md (350 more words)Show less

Implementation checklist (coordinator only)

  1. Keep the MCP tool names and arguments unchanged.
  2. Document the host in SETUP.md with a copy-paste config that uses a resolved agent-bridge absolute path (Get-Command agent-bridge | Select-Object -ExpandProperty Source). Do not invent a home directory.
  3. Update every user-facing coordinator list (README, SETUP, ORCHESTRATION, skill, CLI HELP, package description) so it cannot still say "Codex / Cursor / Kimi only".
  4. If the host can also be a Worker, add or keep the nested-dispatch tests; do not rely on the model "being careful".
  5. If ORCHESTRATION.md is copied as AGENTS.md (Grok does this), keep the English file ≤ 9500 characters. Chinese can stay longer; behavior must match.

Definition of done

Automation:

  • Unit tests cover worker-context refusal (dispatch_task / set_preferences / cancel_task / end_session), nested AGENT_BRIDGE_HOME, skill destinations, and docs that distinguish config shapes.
  • uv run pytest and uv build pass.

Human or live-host:

  • A local lab/ workspace from scripts/setup_lab.py (its own git repo, project-scoped host config; not committed). Do not edit the user's global Codex/Cursor/Kimi/ZCode/Grok/Claude files for the test, and do not use tests/ or a Temp folder.
  • list_agents shows dispatch_enabled=true at the top-level host.
  • Two worker turns on one session_id; the file on disk has both lines.
  • If the host is also a Worker: nested dispatch_task is rejected with nested dispatch is disabled; nested cancel_task / end_session are also rejected; nested state lives under AGENT_BRIDGE_HOME/nested. Top-level end_session still tears down the worker tree and its nested Bridge. Top-level env_status does not count that nested Bridge as a sibling.
  • If the host has no stable headless CLI, say "config and unit tests done; UI E2E waiting on one human prompt". Do not claim full verification.

Common failures

  • Treating a coordinator as a new worker. No adapter, no agents.toml worker block.
  • Mixing UI JSON with native file JSON (ZCode).
  • Mixing seconds and milliseconds.
  • Putting user-only keys in a project example (Grok [ui] permission_mode).
  • Claiming .agents/mcp.json always merges with a native .zcode file. It does not.
  • Calling grok / kimi from the coordinator shell because MCP tools "looked missing".
  • Using user_requested=true to bypass a nested Bridge.
  • Trusting the worker's self-report instead of git diff.

© FeiZhuLulu, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in skills/add-coordinator of FeiZhuLulu/Agent-Bridge.

Open the folder on GitHubat commit 5560e06

Compare with similar skills

Add Coordinator next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Add Coordinator compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Add Coordinator this skillFeiZhuLulu/Agent-Bridge110—~1.9kAutomated safety check: PassMIT
Kimi DatasourceMoonshotAI/kimi-code7.8k—~1.8kAutomated safety check: PassMIT
Perplexity Web MCP Bridgejacob-bd/perplexity-web-mcp198—~7.6kAutomated safety check: PassMIT
MCP Server Builderanthropics/skills180k63 repos~2.3kAutomated safety check: PassApache-2.0
MCP Server BuildershareAI-lab/learn-claude-code78k4 repos~1.2kAutomated safety check: PassMIT
MCP Integration for Pluginsanthropics/claude-plugins-official38k11 repos~3.1kAutomated safety check: PassApache-2.0

Similar skills

  • Kimi Datasource

    MoonshotAI/kimi-code

    Universal data-source assistant for stocks (Wind, S&P, SEC EDGAR), macro (World Bank, IMF, FRED, NBS), Chinese government data and standards (GB/HB/DB/TT), corporate, academic, legal, WHO/FAO/OECD…

    7.8k GitHub stars~1.8k tokensUpdated today
    Agent WorkflowsAuto-check passed

  • Perplexity Web MCP Bridge

    jacob-bd/perplexity-web-mcp

    Searches the web and queries premium AI models through Perplexity's own web interface, tracking quota carefully across quick, Pro and Deep Research tiers.

    198 GitHub stars~7.6k tokensUpdated 6 days ago
    Productivity & AutomationAuto-check passed

  • MCP Server Builder

    anthropics/skills

    Official

    Guides the design and implementation of Model Context Protocol servers in TypeScript or Python, from tool naming and error messages to evaluation.

    180k GitHub starsUsed in 63 repos~2.3k tokens
    Agent WorkflowsAuto-check passed

  • MCP Server Builder

    shareAI-lab/learn-claude-code

    Walks through building MCP servers in Python or TypeScript that expose tools, resources and prompts to Claude, with templates, registration and testing.

    78k GitHub starsUsed in 4 repos~1.2k tokens
    Agent WorkflowsAuto-check passed

  • MCP Integration for Plugins

    anthropics/claude-plugins-official

    Official

    Explains how to bundle Model Context Protocol servers in a Claude Code plugin, covering config files, stdio, SSE, HTTP and WebSocket server types, and authentication.

    38k GitHub starsUsed in 11 repos~3.1k tokens
    Agent WorkflowsAuto-check passed

  • Crush Configuration

    charmbracelet/crush

    Explains how to configure the Crush coding agent with crushrc or crush.json, covering providers, models, LSPs, MCP servers, hooks, permissions and config precedence.

    29k GitHub stars~3.7k tokensUpdated today
    Agent WorkflowsAuto-check passed

More from FeiZhuLulu/Agent-Bridge

  • Add Worker

    FeiZhuLulu/Agent-Bridge

    Connect a new worker agent CLI to Agent Bridge. An agent skill from FeiZhuLulu/Agent-Bridge.

    110 GitHub stars~2.9k tokensUpdated 4 days ago
    Auto-check passed

  • Agent Bridge

    FeiZhuLulu/Agent-Bridge

    Coordinate local worker agents (Grok Build, Kimi Code, Antigravity, DeepSeek Harness, OpenCode, Claude Code, Codex CLI, Devin CLI, ZCode, MiniMax Code) through the Agent Bridge MCP tools.

    110 GitHub stars~2.4k tokensUpdated 4 days ago
    Auto-check passed

Works with

Categories

Questions about Add Coordinator

What does Add Coordinator do?

Verify that a new host can act as an Agent Bridge coordinator over stdio MCP. Add Coordinator is an agent skill from FeiZhuLulu/Agent-Bridge. Verify that a new host can act as an Agent Bridge coordinator over stdio MCP.

When should I use Add Coordinator?

Add Coordinator fits situations like: adding a coordinator host; checking MCP/skill/timeout/permission setup for Codex; A host can also run as a worker and nested dispatch must stay disabled.

How do I install Add Coordinator in Claude Code?

Run `npx skills add FeiZhuLulu/Agent-Bridge --skill add-coordinator -a claude-code`. Or copy the skill folder (skills/add-coordinator in FeiZhuLulu/Agent-Bridge) into .claude/skills/add-coordinator in your project. Claude Code loads it when a task matches its description.

How do I install Add Coordinator in Codex?

Run `npx skills add FeiZhuLulu/Agent-Bridge --skill add-coordinator -a codex`. Or copy the skill folder (skills/add-coordinator in FeiZhuLulu/Agent-Bridge) into .agents/skills/add-coordinator in your project. Codex loads it when a task matches its description.

Can I use Add Coordinator in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add FeiZhuLulu/Agent-Bridge --skill add-coordinator -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/add-coordinator, .gemini/skills/add-coordinator, .github/skills/add-coordinator and .opencode/skills/add-coordinator in your project.

What does Add Coordinator need to run?

Going by SKILL.md and its folder, Add Coordinator needs the command-line tools its instructions call (uv and git).

Does Add Coordinator access the network?

SKILL.md contains no URLs. Its commands use uv and git, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is Add Coordinator safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Add Coordinator use?

Add Coordinator is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Add Coordinator use?

About 1.9k tokens (SKILL.md is roughly 7.6k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Add Coordinator?

Skills that share tags, products or a category with Add Coordinator: Kimi Datasource (MoonshotAI/kimi-code, 7.8k stars), Perplexity Web MCP Bridge (jacob-bd/perplexity-web-mcp, 198 stars), MCP Server Builder (anthropics/skills, 180k stars) and MCP Server Builder (shareAI-lab/learn-claude-code, 78k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Add Coordinator?

FeiZhuLulu (a GitHub user) maintains it in FeiZhuLulu/Agent-Bridge, which has 110 GitHub stars. The repository holds 3 skills in this directory. The repository was last updated on October 7, 2026.

Source: FeiZhuLulu/Agent-Bridge on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.