AIPI Bridge MCP
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@AIPI Bridge MCPList my open GTD tasks"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
AIPI Bridge MCP
SSE-transport MCP server exposing a curated slice of Bot Army to external MCP clients that only speak SSE — specifically AIPI's "Custom MCPs" skill config, which documents SSE as the only currently supported format.
Deliberately narrow, not a generic bridge.* passthrough: this is meant to be
reachable from a third-party vendor's cloud (via ngrok), not a trusted local
dev session, so it exposes exactly three tools instead of letting a caller
invoke arbitrary bridge subjects. For local/trusted MCP clients that want the
full surface (any bridge.* subject, NATS observe, registry queries), use
tools/nats-gateway-mcp instead — stdio transport,
generic bridge_request tool.
Tools
Tool | Purpose |
bridge_chat | Ask Bot Army a question via the ChatResponder (tasks, bots, system status). |
gtd_task_list | List GTD tasks. |
gtd_task_create | Create a GTD task. Appends a |
gtd_task_complete | Mark a task complete. |
gtd_task_search | Search tasks by text, with optional status/context filters. |
internal_docs_query | Semantic search over the internal docs corpus. |
random_roll | Dice roll ( |
Deliberately not exposed: bridge.task.update (a misheard voice command could silently mutate an existing task's fields with no confirmation screen — if wanted later, scope it to status→completed only, i.e. reuse gtd_task_complete's shape rather than open-ended field updates) and bridge.project.* (bridge.project.list is a known-broken responder as of 2026-08-15 — returns empty despite tasks carrying project_ids).
Related MCP server: Hermes MCP Bridge
Setup
Build (from repo root or this directory):
cd tools/aipi-bridge-mcp && npm install && npm run buildRun
NATS_URL=nats://localhost:4222 \ AIPI_BRIDGE_MCP_PORT=39895 \ AIPI_BRIDGE_MCP_TOKEN=<pick-a-token> \ npm startAIPI_BRIDGE_MCP_TOKENis required before tunneling this externally — without it the SSE endpoint is unauthenticated. Server logs a warning on startup if it's unset.SSE endpoint:
GET /sse(requiresAuthorization: Bearer <token>), message endpoint:POST /messages?sessionId=..., health check:GET /health(no auth).
Expose it
Same pattern as everything else externally-reachable in this repo: tunnel via
air.ngrok(reserve a port label, e.g.surface-aipi-mcp), point AIPI's Custom MCP config at the resulting HTTPS URL +/sse, with the bearer token in whatever auth field their SSE MCP config exposes.
Requirements
Node.js 18+
NATS server reachable at
NATS_URL— use prod (nats://localhost:4222) for real bridge responses;bridge.chat/bridge.task.*only have responders on the production cluster, not dev (4223).
Development
npm run build— compile TypeScript todist/npm run dev— run with tsx (no build) for quick iterationVerified end-to-end against dev NATS (
4223) with a throwaway MCP client: SSE handshake,tools/list, andtools/callall round-trip correctly; tool calls correctly report failure (NATS503 no responders) rather than fabricating a success, since nothing listens onbridge.chat/bridge.task.liston the dev cluster. Full happy-path verification needs a run against prod NATS.
This server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceExposes any stdio-based MCP server to the internet via HTTP/SSE transport, enabling remote agents to access MCP tools over a network.16MIT
- FlicenseNot gradedqualityCmaintenanceRead-only MCP bridge exposing Hermes gateway monitoring tools (status, mission control, channels, approvals) to ChatGPT via HTTP/SSE.
- FlicenseNot gradedqualityDmaintenanceMCP server that exposes Zulip REST API tools via SSE, enabling message retrieval, stream/topic listing, draft management, and notifications.
- FlicenseNot gradedqualityDmaintenanceA standalone MCP server that exposes API endpoints as tools for AI assistants, using SSE transport.
Related MCP Connectors
An authenticated remote MCP server for user-owned devices and one-shot capability invocation.
MCP server exposing the Backtest360 engine API as tools for AI agents.
Hosted NeuroDock — stateless communication and planning tools over OAuth-secured Streamable HTTP.
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/the-great-abby/ergon_aipi-bridge-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server