Skip to main content
Glama

run_launch

Start or resume a single version's run over diagnosis or selection tasks, then poll status and read results. Detached by default to measure one arm before comparing candidates.

Instructions

Start or resume one arm: a run of one version over the tasks of the diagnosis or selection split, scored by the scenario. Use it to measure a single version; use run_interleaved to run a parent and candidates side by side. Detached by default: it returns at once while a background worker runs, so poll run_status, then read run_results. Writes run records, results and traces, and costs whatever the scenario's tasks cost (for example model calls). Resuming a run_id with the same parameters only fills missing task slots; changed parameters are refused. Returns the run status object.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "config": "<config>",
      "version": "<version>"
    },
    "name": "run_launch"
  }
}

Input Schema

NameRequiredDescriptionDefault
waitNoBlock until every task finishes instead of returning at once. Default false.
smokeNoRun only the first two tasks of the split as a quick check. Smoke arms cannot be gated or used for noise. Default false.
splitNoTask split: `diagnosis` to explore and read failures, `selection` for gate evidence. Default `selection`. Confirmation and final splits are never available.selection
configYesPath of the scenario YAML file on the server host, absolute or relative to the server's working directory, for example `scenario.yaml`. It selects the workspace; a server started with --config rejects any other path.
run_idNoArm name. Omit for a new random id; pass an existing id with identical parameters to resume that arm. Run ids are 1 to 80 letters, digits, `_` or `-`, starting with a letter or digit.
repeatsNoTimes each task runs, at least 1. Default 1.
versionYesVersion to run: 64-character lowercase SHA-256 version id, as returned by version_create.
workersNoParallel task workers, 1 to 256. Default 1.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed13 schema fields changedv0.1.9
    • addedInput schema / properties / config / description
      Added value: +"Path of the scenario YAML file on the server host, absolute or relative to the server's working directory, for example `scenario.yaml`. It selects the workspace; a server started with --config rejects any other path."
    • addedInput schema / properties / repeats / default
      Added value: +1
    • addedInput schema / properties / repeats / description
      Added value: +"Times each task runs, at least 1. Default 1."
    • addedInput schema / properties / run_id / description
      Added value: +"Arm name. Omit for a new random id; pass an existing id with identical parameters to resume that arm. Run ids are 1 to 80 letters, digits, `_` or `-`, starting with a letter or digit."
    • addedInput schema / properties / smoke / default
      Added value: +false
    • addedInput schema / properties / smoke / description
      Added value: +"Run only the first two tasks of the split as a quick check. Smoke arms cannot be gated or used for noise. Default false."
    • addedInput schema / properties / split / default
      Added value: +"selection"
    • addedInput schema / properties / split / description
      Added value: +"Task split: `diagnosis` to explore and read failures, `selection` for gate evidence. Default `selection`. Confirmation and final splits are never available."
    • addedInput schema / properties / version / description
      Added value: +"Version to run: 64-character lowercase SHA-256 version id, as returned by version_create."
    • addedInput schema / properties / wait / default
      Added value: +false
    • addedInput schema / properties / wait / description
      Added value: +"Block until every task finishes instead of returning at once. Default false."
    • addedInput schema / properties / workers / default
      Added value: +1
    • addedInput schema / properties / workers / description
      Added value: +"Parallel task workers, 1 to 256. Default 1."
  2. First observedv0.1.7

TDQS

A4.6/5.0
Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Adds substantial behavior beyond the annotations: detached-by-default execution with a background worker, polling pattern, the fact that it writes run records, results and traces, and that it incurs real cost from scenario tasks such as model calls. It also discloses resume semantics (only fills missing task slots; changed parameters refused), which the openWorldHint/destructiveHint flags alone would not convey. Minor gap: no mention of concurrency or cost implications of the workers parameter.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Front-loaded with purpose and the sibling comparison, then execution model, side effects, cost, and resume rules in a logical order. Every sentence carries information, though the block is dense and slightly run-on; splitting the detached/polling sentence from the cost sentence would improve scannability without losing content.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a mutating, costly, open-world tool with no output schema, the description covers everything an agent needs: what it does, the alternative, that it returns immediately and must be polled, what it writes, what it costs, resume constraints, and what it returns ('the run status object'). No material gap remains.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so the schema already carries the per-parameter documentation and the baseline would be 3. The description earns above baseline by explaining the resume contract at a behavioral level — an existing run_id with identical parameters only fills missing task slots and changed parameters are refused — which clarifies the run_id parameter's operational meaning rather than restating its type.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource ('Start or resume one arm') and defines the resource precisely: one version over the tasks of the diagnosis or selection split, scored by the scenario. It immediately distinguishes itself from run_interleaved, so an agent can separate the two without opening either schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives explicit selection guidance: 'Use it to measure a single version; use run_interleaved to run a parent and candidates side by side.' It also names the downstream follow-up tools (poll run_status, then read run_results), which tells the agent what to do after invocation rather than leaving it to inference.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.