agency-mcp-server
About
On-demand access to 150+ specialist AI agent templates — search, browse, and spawn agents. 150x reduction in context usage vs loading agents locally.
Details
- Author
- npupko
- Categories
- Developer Tools, AI, Automation
Jump to
Setup
Install agency-mcp-server in your MCP client (Claude Desktop, Cursor, Windsurf, and others).
Repository: https://github.com/npupko/agency-mcp-server
Follow the installation instructions in the repository README, then restart your MCP client.
One MCP config entry. 150+ specialist agents on demand. No manual setup.
Your AI assistant is a generalist. Sometimes you need a specialist -- a game economy designer, a security auditor, a technical writer. This MCP server gives your assistant instant access to 150+ expert agent templates. Describe what you need, it finds the right agent and spawns it.
You: "Help me design a balanced game economy" Claude: [searches -> finds Game Economy Designer -> spawns it -> expert response]
Templates auto-fetch on first run fromagency-agentsand stay updated. You don't touch a thing.
That's a150x reductionin baseline context usage. You get the same 160+ agents, but you only pay for the one you're actually using.
Installed agents (8,300 tokens):We ran the agency-agents install script (install.sh --tool claude-code), which copied 162 agent files to~/.claude/agents/. Then opened a fresh Claude Code session and ran/context. Claude Code reported "Custom agents: 8.3k tokens" -- loaded into every conversation regardless of whether any agent is used.
MCP idle (55 tokens):With the MCP server configured instead,/contextshows only the two deferred tool names (agency_search,agency_browse) and a brief server description in the system prompt. No agent data is loaded.
MCP searching (350 tokens):Measured by tokenizing the full JSON tool schemas that get loaded when the assistant callsToolSearchto resolve theagency_searchandagency_browsetools. Counted with@anthropic-ai/tokenizer.
MCP using an agent (2,700 tokens):The median token count across all 145 agent files, measured with@anthropic-ai/tokenizer. Only the single agent file you're actually using gets loaded into context. The range is 383–12,724 tokens depending on the agent (p25: 1,549, p75: 3,584).
/plugin marketplace add npupko/agency-mcp-server /plugin install agency@agency-mcp-server
claude mcp add agency -- npx -y agency-mcp-server
{ "mcpServers": { "agency": { "command": "npx", "args": ["-y", "agency-mcp-server"] } } }
That's it. First launch clones templates to~/.cache/agency-mcp-server/and pulls updates every 24 hours.
"Search for a game economy designer agent"
You should see results from theagency_searchtool. If it's the first run, templates will auto-download (~30 seconds).
- agency_search(query, division?)-- describe a task, get matching agents with spawn instructions
- agency_browse(division?)-- explore divisions and agents when you want to see what's available
- agency_status()-- check index freshness: agent count, last update time, whether an update is available
- agency_update()-- pull latest templates from git and rebuild the search index without restarting
When you ask for help with something specific, your assistant callsagency_search, picks the best match, and spawns a subagent with that specialist's full system prompt. You get an expert response without ever touching a config file.
All configuration is through environment variables in your MCP config:
{ "mcpServers": { "agency": { "command": "npx", "args": ["-y", "agency-mcp-server"], "env": { "AGENCY_AGENTS_PATH": "/path/to/your/agent-templates" } } } }
{ "mcpServers": { "agency": { "command": "npx", "args": ["-y", "agency-mcp-server"], "env": { "AGENCY_REPO_URL": "https://github.com/yourorg/custom-agents.git" } } } }
Each agent is a Markdown file with YAML front-matter, organized by division:
engineering/ software-architect.md devops-engineer.md design/ ui-designer.md game-development/ game-economy-designer.md
--- name: Software Architect description: Expert software architect specializing in system design... --- Full agent system prompt goes here.
The server indexes thenameanddescriptionfields for search. The full Markdown body becomes the agent's system prompt when spawned.
- agency_search(query, division?)-- find agents by task description, returns matches with file paths and a ready-to-use spawn template
- agency_browse(division?)-- list all divisions, or list agents within a specific division
- agency_status()-- check index freshness: agent count, last update time, whether an update is due
- agency_update()-- pull latest templates from git and rebuild the search index at runtime
- agency://agents-- full agent index as JSON
- agency://divisions-- division list with counts and examples
- use-agent-- describe a task, get the best-matching agent with spawn instructions
npm install npm run build # Run with auto-fetched templates node dist/index.js # Run with local templates AGENCY_AGENTS_PATH=./my-agents node dist/index.js # Type checking npm run typecheck # MCP Inspector npm run inspect
Agent templates fromagency-agentsby@msitarzewski.
This is a web browser that enables your coding agent, such as Claude Code, to visit websites on your behalf and assist you in identifying bugs or creating UI test cases.
What Shopify did for ecommerce, Chipp does for AI agents. Build, deploy, and monetize AI agents for your business — no engineering team required.
CodeVF MCP lets AI hand off problems to real engineers instantly, so your workflows don’t stall when models hit their limits.
Client implementation for Mastra, providing seamless integration with MCP-compatible AI models and tools.
Agent-native developer Q&A API with MCP + A2A endpoints for citations, job pickup, and answer submission.
AgentChatBus is a persistent AI communication bus that lets multiple independent AI Agents chat, collaborate, and delegate tasks — across terminals, across IDEs, and across frameworks.
Remote MCP server (Streamable HTTP) at https://mcp.agenticrail.nz/ — deterministic step-order enforcement for AI agents. evaluate_step returns ALLOW or DENY before a step runs; verify_receipt proves a sequence's Ed25519-signed, hash-chained receipt chain is intact. No auth required: omit the bearer token and calls run on the public demo key. That first clause matters — the form has no "remote/hosted" field, and putting the endpoint in the description is the convention on that list ("Fully REMOTE! Just use…"). The rest mirrors your own server card verbatim, so the listing and the card can't drift.
An AI Agent with optional Human-in-the-Loop Safety and Model Context Protocol (MCP) integration.
Agent Mail Rooms is a paid remote MCP endpoint for OpenAI Codex agent coordination MCP. It exposes structured JSON tools, a public server card, token-based access, usage receipts, and audit-
MCP bridge that lets Claude Code delegate heavy tasks to the Antigravity CLI (agy) — purpose-built tools, model routing with fallback, session continuity, and output truncation to save Claude's context and tokens.
A timeline tool for AI agents to post their thoughts and progress while working.
Sign in to leave a review
Use Google, GitHub, or an email account so ratings stay tied to real people.
No reviews posted yet.
