Launch Imminent — Warden v0.1.0
A lean, multi-provider, terminal-native LLM coding agent. Single binary. MIT. Pre-Alpha ships soon.
Something is coming.
A terminal-native LLM coding agent that doesn’t lock you to one provider.
Anthropic Claude, OpenAI (Platform API + Responses subscription tier), and
Google Gemini — all fully wired today. Bedrock, Vertex, and Foundry
scaffolded for tomorrow, each raising explicit NotImplementedError so you
can’t accidentally use them.
Twenty-one builtin tools. A full MCP client and server mode. Six permission
modes with an eight-source precedence chain. Twenty-two hook event types. A
13-tier import DAG enforced by import-linter on every PR.
42,800 lines of Python 3.14. MIT licensed. Single binary. Telemetry is off by
default — opt-in for full diagnostic, env-killable via WARDEN_TELEMETRY=off,
schema-allowlisted server-side. The license activation flow uses a small
first-party endpoint at api.wardenclient.com; the full data contract is
documented in docs/telemetry.md in the repository.
We don’t promise vaporware. The Pre-Alpha tag is honest — three providers fully wired, three scaffolded with explicit error placeholders, the known gaps documented on the homepage of the README itself. What ships, ships.
Stay tuned.