Log inSubmit server
Honcho logo

Honcho

Adds persistent, reasoning-based memory to agents through Honcho’s managed or self-hosted service.

InstallRepositoryWebsite
Full README Compare alternatives Embed widget Claim this listing

Install Honcho

Pick your client, then add it in one click or paste the config.

STDIOPython
~/Library/Application Support/Claude/claude_desktop_config.json
claude_desktop_config.json
{
  "mcpServers": {
    "honcho": {
      "command": "uvx",
      "args": [
        "honcho-ai"
      ],
      "env": {
        "HONCHO_API_KEY": "YOUR_VALUE_HERE",
        "HONCHO_URL": "YOUR_VALUE_HERE"
      }
    }
  }
}

Settings → Developer → Edit Config, merge this into mcpServers, then fully quit and reopen Claude Desktop.

Values are written into the snippet above so you can paste it as is. They never leave your browser.

We couldn’t confirm it starts correctly

We ran the install command but it didn't respond within our test window — this can mean a slow first-time install rather than a real problem.

uvx honcho-ai

Process exited before answering initialize.

This is an experimental automated check and can have false negatives — missing environment variables, a slow cold install, etc. It doesn’t necessarily mean something’s wrong. Last checked Today.

Overview

The Honcho MCP server connects agents to Honcho’s memory infrastructure for storing conversations and events, then retrieving context about peers, sessions, and workspaces. Honcho processes stored information asynchronously and supports chat, search, representations, prompt-ready context, document uploads, and queue inspection. Use it when an agent needs memory that persists across interactions and can reflect changing users, agents, groups, projects, or ideas. It can use the managed Honcho API, a local CLI-launched stack, or a self-hosted FastAPI deployment.

Use cases

  • Store agent conversations for later sessions
  • Retrieve user or project context before generating a response
  • Search workspace memory with hybrid retrieval
  • Share memory across supported coding agents

Key features

  • Persistent session messages and events
  • Peer and workspace natural-language queries
  • Hybrid BM25 and vector search
  • Asynchronous background reasoning
  • Prompt-ready context conversion
  • Managed, local, or self-hosted deployment

Security & health

What it can reach and how reliably it runs — from our own checks, not the maintainer’s claims.

ActiveLast checked 2d ago

Recent health check succeeded.

10/10 checks healthy over the last 39d

Supply-chain signal

No high-severity advisories surfaced by our automated scan.

Critical 0High 0Medium 0Low 0

Scanned 11d ago via OSV.dev · honcho-ai (PyPI)

Adoption & maintenance

Factual signals from GitHub, npm, and our automated checks — not a rating.

GitHub stars7.7kStargazers on the source repository.
Last commit2d agoMost recent push to the default branch.
Install checkInconclusiveDidn't respond in our test window — often a slow first install.
Directory activity1 viewsConfig copies, upvotes, and views on AllMCPs.

How Honcho works

What the Honcho MCP server does

The Honcho MCP server gives an MCP-compatible agent access to Honcho’s stateful memory model. It stores messages, events, documents, and tool traces in sessions, associates activity with peers, and makes accumulated information available for later queries. Peers can represent users, agents, groups, projects, or ideas whose knowledge and state change over time.

Rather than treating memory only as a collection of matching text fragments, Honcho builds representations from stored activity and processes new information in the background. Agents can request what Honcho knows about a peer, ask questions across a workspace, retrieve session context for a prompt, or search stored material.

The Honcho MCP server is relevant for coding agents and other MCP clients that need continuity between conversations. It is also suitable when the same memory should be shared across multiple agent integrations using a common Honcho configuration.

How it works

A typical flow has four stages:

  1. Add conversations, events, documents, or tool traces to a session.
  2. Let Honcho process the new information asynchronously.
  3. Query a peer, session, or workspace for context, search results, representations, or a natural-language answer.
  4. Insert the returned context into an LLM or agent request.

Workspaces contain peers, peers participate in sessions, and sessions contain messages. Honcho can model relationships between peers when configured, including what one peer knows about another. Search supports hybrid BM25 and vector retrieval through peer, session, and workspace queries.

The API also provides static peer and session representations for lower-latency reads. Newly added messages may not appear immediately in chat or representation results because background reasoning runs asynchronously. The representation endpoint is recommended in the README for low-latency reads.

Setup and configuration

Managed deployments are available through api.honcho.dev. Obtain an API key from app.honcho.dev, then configure the client with a workspace ID and HONCHO_API_KEY. The Python SDK package is honcho-ai; the TypeScript package is @honcho-ai/sdk.

For local use, install the Honcho CLI and run honcho start --setup, then direct the SDK to http://localhost:8000. Self-hosting from source is also supported through Docker Compose or local development. Set HONCHO_URL when the service is not using the managed default.

The project’s integrations use a shared ~/.honcho/config.json, allowing one configuration to be used across supported agent integrations.

Tools and capabilities

Documented Honcho capabilities available to an agent include:

  • Adding messages to a session for persistent interaction history.
  • Asking a peer or an entire workspace natural-language questions.
  • Streaming workspace chat responses.
  • Searching peers, sessions, or workspaces with hybrid retrieval.
  • Fetching peer or session representations for faster reads.
  • Building prompt-ready session context for OpenAI or Anthropic message formats.
  • Checking background processing with queue status.

These capabilities can be used with any model or framework that accepts the resulting context. Honcho itself does not replace the downstream model call; the retrieved context is handed to the model or agent by the application.

Limitations and notes

Background reasoning is asynchronous, so recent messages can take time to influence chat and representation responses. Low-latency applications should use the representation endpoint where appropriate.

Confirm those details in the repository before deploying a client-specific configuration.

Managed, local, and self-hosted deployments have different service URLs. A workspace ID is required by the SDK examples, and managed use requires an API key. Local deployments can instead target the local service URL.

Read the full README View source on GitHub

Reviews

No reviews yet — be the first to share how it worked for you.

Frequently Asked Questions about Honcho

Honcho is an MCP server that connects agents to Honcho’s memory infrastructure. Its main capabilities include storing session messages, querying peers or workspaces, searching memory, retrieving context and representations, uploading files, and checking background processing. It supports managed, local, and self-hosted Honcho deployments.

Install and configure Honcho through the repository’s documented integration or run the Honcho CLI with `honcho start --setup` for a local stack.

Yes, the README identifies Cursor-compatible clients among Honcho’s agent-tool integrations.

Set the service URL to your self-hosted Honcho instance, using `HONCHO_URL` or the SDK’s `base_url` option.

Related MCP servers

All in Knowledge & Memory All alternatives
Add a badge to your READMEShow it’s listed — live SVG, Markdown or HTML.

AllMCPs Directory Badge

Full Badge Customizer

Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.

Badge Style:
Live Dynamic SVG PreviewHoncho AllMCPs Directory Badge
Markdown (GitHub README)
[![AllMCPs](https://allmcps.com/api/badge/honcho?style=directory)](https://allmcps.com/mcp/honcho)
HTML Embed
<a href="https://allmcps.com/mcp/honcho"><img src="https://allmcps.com/api/badge/honcho?style=directory" alt="Honcho on AllMCPs" /></a>