mechbench-runner

The machine-side process of the mechbench family: it claims queued jobs from mechbench-api, executes them against mechbench-compute, and posts results back. It also exposes those same primitives as Model Context Protocol tools, so an LLM agent can call them directly.

Status: in use. login pairs a machine with an account; the runner then claims and executes jobs, reports progress and preparing steps, holds a live WSS channel for control and telemetry, and installs as a launchd or systemd service so it survives reboots. doctor tells you whether a machine will work before it tries. Three MCP tools (run_protocol, get_result, list_jobs) expose the same primitives to an agent.

What this repo is for

Two adjacent surfaces for different callers:

1. MCP server. An LLM agent (Claude, others) connects via MCP stdio and calls mechbench primitives as structured tools. Tool bodies run in-process against mechbench-compute.
2. Job-runner. Polls mechbench-api's /jobs/next for UI-queued protocols, runs them, posts results back. Same compute path as the MCP run_protocol tool; different trigger.

Both modes share one binary (mechbench-runner) with subcommands; they share the loaded model, API client, and protocol executor. Splitting into separate processes is a later operational decision — see "Open design questions" below.

Architectural decisions (task 000185)

• Python. mechbench-compute is Python; delegating to Python via RPC or subprocess-shell from a TS runner adds a layer that pays no dividends in v0. The MCP Python SDK is mature.
• One binary, two subcommands. mechbench-runner mcp launches the MCP server over stdio; mechbench-runner run starts the job-runner loop. They share ExperimentRunner (owns the loaded Gemma model) and ApiClient.
• Agent authenticates to mechbench-api with a dedicated API key, not a user's personal session. Export MECHBENCH_API_KEY (mint one at /settings/api-keys, or via POST /auth/api-keys). Matches the pattern from the e2e trace.
• MCP run_protocol runs in-process, not queued through mechbench-api. The MCP caller wants the answer; we are the compute target. Job-queue round-tripping exists for the UI-triggered path (job-runner subcommand).
• stdio transport only. SSE / HTTP-SSE transports earn their seat once remote MCP deploy matters (deferred).

Install

uv tool install mechbench-runner     # or: pipx install mechbench-runner
mechbench-runner login

login prints a link, takes the registration token from it, stores a durable key at ~/.mechbench/config.toml (mode 0600), and offers to start the runner automatically. Say yes and there is nothing further to do: it starts at login, comes back after a crash, and is controlled from the website.

mechbench-runner doctor answers "will this actually work here" — Python, backend, credentials, API, model cache, disk — before you find out the slow way.

Running a model needs Apple Silicon (the MLX backend from mechbench-compute). The rest installs anywhere.

Running it yourself

mechbench-runner run              # foreground, ^C to stop
mechbench-runner install-agent    # or have the OS keep it running
mechbench-runner agent-status

The service is supervised by launchd or systemd rather than by anything we wrote — see mechbench_runner/exits.py for the contract that makes that work.

From a checkout

git clone https://github.com/mechbench/mechbench-runner.git
cd mechbench-runner
python3.11 -m venv .venv
source .venv/bin/activate
pip install -e '.[dev]'

Usage

MCP server

Launch as a stdio MCP server — connect from Claude Desktop via claude_desktop_config.json:

{
  "mcpServers": {
    "mechbench": {
      "command": "/abs/path/to/mechbench-runner/.venv/bin/mechbench-runner",
      "args": ["mcp"],
      "env": {
        "MECHBENCH_API_URL": "http://localhost:3000",
        "MECHBENCH_API_KEY": "mbk_..."
      }
    }
  }
}

Three tools appear in Claude:

Job-runner

Polls mechbench-api for UI-queued jobs. Same compute path as run_protocol; different trigger.

export MECHBENCH_API_URL=http://localhost:3000
export MECHBENCH_API_KEY=mbk_...
mechbench-runner run

Ctrl-C exits cleanly. API-unreachable is retried with exponential backoff capped at 30 s.

In-process smoke test

mechbench-runner smoke            # quick: list_jobs + get_result
mechbench-runner smoke --full     # adds run_protocol (42 forwards, ~1-2 min)

Configuration

All via env vars:

Relationship to other mechbench repos

• mechbench-compute — imported directly. Model, Ablate, hook-aware forward.
• mechbench-schema — produces LayerAblationPayload etc. as typed results.
• mechbench-api — the runner's only platform dependency. All workspace state (jobs, cache reads) goes through it.
• mechbench-ui — no coupling. UI queues jobs; the job-runner consumes them.
• mechbench-experiments — research scripts that use mechbench-compute directly, without the job machinery.

Open design questions (deferred)

• One binary or two processes? Current answer: one binary, two subcommands. Revisit if MCP-caller frequency vs. job-runner throughput diverges enough to want independent scaling.
• Structured-summary interface. The family's philosophy doc describes a read-side surface where agents consume JSON summaries of findings / experiments. Currently implicit in list_jobs + get_result. A richer summary layer (GET /summary, POST /query) is still on the table but unbuilt.
• MCP-surface observability. Rate limits, per-tool metrics, audit trail for the tool-calling side. Deferred until a second LLM-agent consumer exists.

License

MIT.