oto — CLI toolkit for AI agents

Your LLM already uses gh for GitHub, aws for AWS, gcloud for GCP. oto covers the long tail — the SaaS products that don't have a CLI.

The CLI is self-documenting via --help, so your AI agent discovers every connector on its own. For Claude Code, the oto plugin (otomata-tech/oto-plugin) bundles a universal skill + the Oto MCP connector.

Why CLI over MCP?

Installation

# Standalone CLI
pipx install oto-cli

# With specific connectors
pipx install "oto-cli[google,browser]"

# All connectors
pipx install "oto-cli[all]"

# Development
git clone https://github.com/otomata-tech/oto-cli.git
cd oto-cli && pip install -e ".[all]"

Connectors

Connectors without an "Extra" only need requests (included in base install).

Quick start

# Configure secrets (file-based, default)
mkdir -p ~/.otomata
cat > ~/.otomata/secrets.env << 'EOF'
SERPER_API_KEY=xxx
NOTION_API_KEY=secret_xxx
SIRENE_API_KEY=xxx
EOF

# Or use Scaleway Secret Manager
oto config provider secrets scaleway

# Check config & secrets status
oto config

# Search the web (backend depends on config: serper or browser)
oto search web -q "AI agents 2026"

# Google OAuth setup (per account)
oto google auth myaccount
oto google gmail-search "from:bob" -a myaccount

# Browse LinkedIn
oto browser linkedin profile https://linkedin.com/in/someone

# French company data — INSEE SIRENE
oto sirene search "fintech"

# Enriched data via data.gouv recherche-entreprises (with --idcc filter by convention collective)
oto sirene entreprises --idcc 1285,3090 --naf 90.01Z --dept 75

For AI agents

The CLI is self-documenting — agents discover everything via --help:

oto --help                  # list namespaces
oto <namespace> --help      # commands in a namespace
oto <namespace> <cmd> --help   # arguments

For Claude Code, the oto plugin (otomata-tech/oto-plugin) bundles a single universal skill (entry-point doctrine) plus the Oto MCP connector — install it once instead of per-tool manuals.

Create your own connector

A connector is 2 files:

oto/commands/myservice.py    # CLI commands (Typer app)
oto/tools/myservice/         # API client

See docs/create-connector.md for the full guide.

How it works

Auto-discovery — cli.py scans commands/ at startup. Any Python file exporting a Typer app becomes a sub-command. No registry, no config. Drop a file, it appears in oto --help.

Three types of connectors (same CLI contract: JSON stdout, lazy imports, get_secret()):

• API — call REST APIs via requests. Each client handles auth/pagination/rate-limiting as the API requires. (notion, sirene, search, pennylane, folk...)
• Browser — automate a real browser for sites without an API. Async, require oto-cli[browser]. (linkedin, crunchbase, pappers, indeed, g2)
• SDK — use an official client library. Currently Google Workspace via google-api-python-client + OAuth2. Require oto-cli[google].

Secrets — provider-based resolution, configured via oto config provider secrets <file|scaleway>:

1. Environment variables (always, highest priority)
2. Configured provider: file (.otomata/secrets.env project → user) or Scaleway Secret Manager
3. Default value

Output contract — every command prints JSON to stdout, errors to stderr. Composable with jq, pipes, scripts.

Lazy imports — tool clients are imported inside functions, so the CLI starts fast regardless of how many connectors are installed.

See docs/concepts.md for the full architecture guide.

License

MIT