██╗  ██╗ ██████╗ ███╗   ██╗ ██████╗██╗  ██╗ ██████╗
██║  ██║██╔═══██╗████╗  ██║██╔════╝██║  ██║██╔═══██╗
███████║██║   ██║██╔██╗ ██║██║     ███████║██║   ██║
██╔══██║██║   ██║██║╚██╗██║██║     ██╔══██║██║   ██║
██║  ██║╚██████╔╝██║ ╚████║╚██████╗██║  ██║╚██████╔╝
╚═╝  ╚═╝ ╚═════╝ ╚═╝  ╚═══╝ ╚═════╝╚═╝  ╚═╝ ╚═════╝

honcho-cli

A terminal for Honcho — memory that reasons.

Install

As a standalone tool (recommended):

uv tool install honcho-cli

Quick Start

honcho init                 # confirm/set apiKey + Honcho URL in ~/.honcho/config.json
honcho start --setup basic  # local stack: LLM provider key + Docker
honcho doctor               # verify config + connectivity
honcho install              # find your AI agents and add Honcho to them
honcho                      # show banner + command list

honcho init writes apiKey and environmentUrl to the top-level of ~/.honcho/config.json (the same file other Honcho tools — plugins, host integrations — share) so the CLI can call a Honcho server. If both are present, it confirms them with you; if either is missing (or you decline), it prompts and writes them back. Host-specific entries under hosts are left untouched. It does not set the LLM provider key the local deriver needs — that is honcho start --setup (or LLM_*_API_KEY in the environment).

Per-command scoping (workspace / peer / session) is handled via -w / -p / -s flags or HONCHO_* env vars — not persisted as CLI defaults.

Local stack

honcho start --setup basic runs a personal Honcho server on your machine (API, deriver, Postgres, Redis) via Docker. The wizard writes the LLM provider key into the profile .env — honcho init cannot do this; its apiKey is for calling a Honcho server, not for deriver/dialectic inference. You can also pass LLM_OPENAI_API_KEY, LLM_ANTHROPIC_API_KEY, or LLM_GEMINI_API_KEY in the environment and skip --setup. Stack files live under ~/.honcho/profiles/local/ and are not committed to a project.

On first start, the CLI pulls ghcr.io/plastic-labs/honcho:latest and pins that digest in profile.json, then copies the image's config.toml.example to config.toml in the same directory. honcho start never overwrites config.toml after that — including when you re-pin the image. Delete the file yourself if you want a fresh copy from a new image.

Pass --setup basic or --setup advanced for an interactive wizard that writes curated LLM/feature overrides into the profile .env (env wins over config.toml). TTY only; re-runnable. basic asks provider + chat model; advanced also covers embeddings, deriver/dialectic models, dreams, and snappy deriver flush. Everything else stays in config.toml.

honcho start does not change environmentUrl in ~/.honcho/config.json (that file is shared with plugins). To talk to the local stack for one command:

HONCHO_BASE_URL=http://127.0.0.1:8000 honcho workspace list

To make local the default, run honcho init --base-url http://127.0.0.1:8000.

honcho start --setup basic
honcho start --setup advanced
honcho status
honcho stop            # keep data
honcho stop --wipe     # also delete volumes

Commands

Onboarding

Workspaces

Peers

Sessions

Scopes

Messages

Conclusions (observations)

Config

Agent Usage

All commands output JSON when stdout isn't a TTY, or when --json is forced. Collection commands emit JSON arrays, and single-resource commands emit JSON objects:

honcho peer list --json
honcho workspace inspect --json | jq '.peers'
honcho doctor --json              # machine-parseable health checklist

Errors are structured:

{
  "error": {
    "code": "PEER_NOT_FOUND",
    "message": "Peer 'abc' not found in workspace 'my-ws'",
    "details": {"workspace_id": "my-ws", "peer_id": "abc"}
  }
}

Non-interactive onboarding:

# Pre-seed via flags / env vars; init still prompts for anything missing
HONCHO_API_KEY=hch-v3-xxx honcho init --base-url https://api.honcho.dev

Agent skill

honcho-cli ships with a skill that teaches agents the right commands and conventions for inspecting and debugging a Honcho deployment. Install it anywhere skills are accepted (Claude Code, other skill-aware agents):

npx skills add plastic-labs/honcho

The picker lists every skill for Honcho — select honcho-cli .

Environment Variables

All HONCHO_* env vars work at runtime — no config file required.

Precedence (highest first): flag → env var → config file → default.

# Per-command flags
honcho peer card -w prod -p user

# Or export once per shell
export HONCHO_WORKSPACE_ID=prod
export HONCHO_PEER_ID=user
honcho peer card

# One-off against a different server
HONCHO_BASE_URL=http://localhost:8000 honcho workspace list

# CI/CD — env vars only, no config file needed
export HONCHO_API_KEY=hch-v3-xxx
export HONCHO_BASE_URL=https://api.honcho.dev
honcho workspace list

Configuration

The CLI shares ~/.honcho/config.json with sibling Honcho tools. It owns two top-level keys: apiKey and environmentUrl (the full Honcho API URL, e.g. https://api.honcho.dev or http://localhost:8000). Everything else at the top level — hosts, sessions, saveMessages, sessionStrategy, etc. — is left untouched.

{
  "apiKey": "hch-v3-...",
  "environmentUrl": "https://api.honcho.dev",
  "hosts": { "claude_code": { "...": "..." } }
}

workspace_id / peer_id / session_id are per-command only — never persisted to the config file.

Development

Install from source in editable mode so changes are picked up live:

git clone https://github.com/plastic-labs/honcho
cd honcho
uv tool install --force --editable --from ./honcho-cli honcho-cli

Re-run any time — changes to honcho-cli/src/ are reflected immediately without reinstalling.

License

MIT