miniouto

A minimal, file-driven CLI agent harness built on coreouto.

Three principles:

1. Minimalism — No bloat. Extend with styles.
2. Automation-friendly — Full CLI; TUI is optional.
3. Fluidity — Adapts to any environment.

Preview

Install

uv tool install miniouto
# or
pip install miniouto

License

Apache License 2.0

Quick start

# Either add a custom provider by hand:
miniouto provider custom add \
  --name openai \
  --api-key sk-... \
  --base-url https://api.openai.com/v1 \
  --default-model gpt-5.5            # optional; used when chat --model is not given

# Or browse the lma (llm-model-api) catalog and add from there:
miniouto provider providers          # see all 144 catalog providers
miniouto provider add Anthropic --api-key sk-ant-...   # adds it with the first catalog model

miniouto provider default openai
miniouto style set default
miniouto chat "hello"
miniouto                              # TUI mode

Computer use (GUI control)

The outo agent can operate GUI applications inside a virtual headless display (via py-Vwayland): launch apps, take screenshots, click, type, scroll, drag. Enabled automatically on Linux x86_64 (glibc ≥ 2.28) — no setup needed; the tool simply doesn't appear on unsupported platforms. Set MINIOUTO_COMPUTER=0 to disable it explicitly. Requires a vision-capable model so the agent can see its screenshots.

Model resolution

The active model is chosen by the first match in:

1. miniouto chat --model <name> (per-call override)
2. settings.model (legacy per-session override; cleared whenever the TUI model picker saves)
3. provider.default_model (set by provider add --default-model, provider custom add --default-model, or the TUI model chip)
4. error — no model can be inferred

The subagent can run on a different provider/model. Its provider is chosen by:

1. miniouto chat --subagent-provider <name> (per-call override)
2. settings.subagent_provider (set via miniouto subagent set --provider)
3. inherit outo's provider

Its model is chosen by:

1. miniouto chat --subagent-model <name> (per-call override)
2. settings.subagent_model (set via miniouto subagent set --model)
3. the subagent provider's default_model
4. inherit outo's model (only when no subagent provider is set; otherwise error)

A model-only override (no subagent provider) is also valid — the subagent then runs outo's provider with the overridden model.

The subagent's reasoning effort resolves as: --reasoning flag (per-call, applies to both outo and subagent) > settings.subagent_reasoning (via miniouto subagent set --reasoning) > the subagent provider's reasoning_effort.

Commands

chat flags

Storage

Everything lives under ~/.miniouto/:

• providers.toml — provider configs (one top-level TOML table per provider: api_format, base_url, api_key, default_model, source)
• settings.toml — active provider, model (legacy), style, session, theme, subagent_provider, subagent_model, subagent_reasoning
• style/<name>.md — style documents
• sessions/<name>.json — conversation history
• subagent-results/ — one markdown file per finished background subagent job (the completion alarm names the file)
• logs/ — reserved (currently unused)

Skills live outside this tree, at ~/.agents/skills/<name>/SKILL.md (the Anthropic convention).

Override the root with the MINIOUTO_HOME environment variable.