# plastic-labs/honcho

Memory library for building stateful agents

Repository: https://github.com/plastic-labs/honcho
Canonical: https://ross.abutalabs.com/products/honcho
Homepage: https://docs.honcho.dev
Language: Python
License: AGPL-3.0
License Family: copyleft
Topics: ai, llm, memory, personalization, embeddings, rag, agent-memory, ai-agents, ai-memory, context-engineering, continual-learning, langchain, long-term-memory, python, state-management, vector-database, anthropic, fastapi, openai, typescript
Last push: 2026-08-26T21:55:15+00:00

## Health v2 (maintenance only)
Score: 72/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 35, longevity 77
- inputs: {"age_days": 1088, "days_push": 7, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 6861, forks 852 (observed 2026-08-28T04:09:50.914215+00:00)

## What it is
Honcho is an open-source memory infrastructure service for building stateful AI agents that maintain understanding of users, agents, groups, and ideas over time. It provides a FastAPI server with Python and TypeScript SDKs, an MCP server, and a CLI, usable as a managed API or self-hosted with PostgreSQL/pgvector.

## Use cases
- add persistent long-term memory to ai agents
- give my coding agent memory across sessions
- build a chatbot that remembers users
- store and query agent conversation history
- self-host agent memory infrastructure
- add memory to claude code or cursor via mcp
- track how users change over time for personalization

## When to choose
- you need agents that maintain context and user understanding across sessions
- you want a managed or self-hosted memory layer that works with any LLM or framework
- you want MCP-based memory for coding agents like Claude Code or Cursor
- you need reasoning-derived insights about users, not just raw message storage

## When to avoid
- you only need simple key-value session state without semantic reasoning
- you want a lightweight in-process memory store with no server dependency
- your project requires a permissive license (AGPL-3.0)
- you cannot provide an LLM provider key, since memory extraction requires one

## Facets
- artifact type: service
- maturity: active
- function: rag, vector-database, agent-framework, mcp, state-management, api-framework, sdk
- domain: artificial-intelligence, large-language-models, developer-tools, chatbots
- platform: python, self-hosted, cloud, cli
- tags: agent-memory, long-term-memory, continual-learning, personalization, context-engineering, fastapi, pgvector, embeddings, langchain, mcp-server, ai-agents, retrieval-augmented-generation, nodejs, docker

## Member repositories
- plastic-labs/honcho (main) score 72

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:50.914215+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:41:43.101753+00:00, confidence not recorded.
  - readme: https://github.com/plastic-labs/honcho (fetched 2026-08-28T04:09:50.914215+00:00, sha 4cba9a97a221)
  - homepage: https://docs.honcho.dev (fetched 2026-08-29T08:37:45.504657+00:00, sha 9f9ec46cfc93)
  - site_page: https://honcho.dev/docs (fetched 2026-08-29T08:37:45.509014+00:00, sha 9f9ec46cfc93)
  - site_page: https://honcho.dev/docs/v3/guides/overview (fetched 2026-08-29T08:37:45.510855+00:00, sha 9b2835a72f51)
  - site_page: https://honcho.dev/docs/v3/contributing/self-hosting (fetched 2026-08-29T08:37:45.512922+00:00, sha 1b41463883ec)
  - site_page: https://honcho.dev/docs/v3/api-reference/introduction (fetched 2026-08-29T08:37:45.515157+00:00, sha 56bdd146248c)
  - site_page: https://honcho.dev/docs/changelog/introduction (fetched 2026-08-29T08:37:45.516884+00:00, sha dbd0a97cf3e3)
  - site_page: https://honcho.dev/docs/v3/documentation/introduction/quickstart (fetched 2026-08-29T08:37:45.520788+00:00, sha 2b6304aa9ac3)
  - site_page: https://honcho.dev/docs/v3/documentation/introduction/vibecoding (fetched 2026-08-29T08:37:45.522757+00:00, sha 4b9941953266)
- Data as of 2026-08-30T08:39:29.467469+00:00.