Install Octopus. Run the whole product.

Start the local agent app, give the brain one Goal, and let tentacles supply Feed while the harness records evidence.

1. Install

curl -fsSL https://dangozhang.github.io/Octopus/install.sh | sh
cargo install --git https://github.com/dangoZhang/Octopus octopus-core --locked --bin octopus --force
octopus download

This installs the Rust kernel and bundled product app. download prints the install, update, source archive, and docs links for the current build. The same manifest is published at download.json.

2. Launch

octopus start --check

The launcher prepares state, seed tentacles, the profile registry, three beats, app evidence, and returns to the shell.

3. Run the first loop

octopus first-run "make this repo easier to use"
octopus chat "make setup clearer"
octopus pet desktop
octopus start --open

User input changes Goal. Need, route choice, provider routing, repair, and harness evolution stay inside the agent or evidence surfaces.

4. Check evidence

octopus start --check
octopus doctor
octopus preflight

Use these when preparing a release or diagnosing a machine. They prove the product path can run locally.

Field evolution preview

Run one worker slot from the peer field pool. Octopus writes a Need, lets the editable field harness work, records Feed plus verifier signal, and keeps the pet as a state observer.

octopus evolve parallel --workers 1 --open "advance the peer field objectives toward v0.2.0"
octopus fields summary

Optional model setup

Use Codex login, direct API keys, local OpenAI-compatible servers, or a gateway. Provider setup is runtime plumbing; clean Need text should stay provider-free.

codex login
export OCTOPUS_LLM_BACKEND=codex
export OCTOPUS_LLM_CODEX_COMMAND=codex
octopus provider status
octopus first-run --live "make this repo easier to use"

export OCTOPUS_LLM_BACKEND=openai-compatible
export OCTOPUS_LLM_MODEL=gpt-4.1-mini
export OCTOPUS_LLM_BASE_URL=https://api.openai.com/v1
export OCTOPUS_LLM_API_KEY=<api-key>
octopus provider check