Skip to main content
Glama

start_worker

Starts an autonomous Claude Code worker in a tmux session to work on a specific feature, enabling parallel coding sessions with optional custom prompts.

Instructions

Start a Claude Code worker in a tmux session to work on a specific feature. The worker will run autonomously.

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": {
      "projectDir": "<projectDir>",
      "featureId": "<featureId>"
    },
    "name": "start_worker"
  }
}

Input Schema

NameRequiredDescriptionDefault
featureIdYesID of the feature to work on (e.g., 'feature-1')
projectDirYesAbsolute path to the project directory
customPromptNoOptional custom prompt to give the worker additional context

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.1.0

TDQS

B3.4/5.0
Behavior3/5

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

With no annotations supplied, the description carries the full behavioral burden. It does disclose meaningful traits — execution happens inside a tmux session and the worker proceeds autonomously without further input — but it omits whether the call blocks or returns immediately, what identifier comes back for later polling with check_worker, whether a session already running causes a conflict, or what permissions/environment the worker needs.

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

Conciseness5/5

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

Two short sentences with zero filler; the core action and its mechanism are front-loaded in the first clause before the autonomy note. Nothing here is redundant.

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

Completeness2/5

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

There is no output schema and no annotations, so the description is the only source of behavioral detail for a tool that spawns a long-running background process. It leaves out the return value, how the agent is expected to monitor the worker afterwards, and failure/conflict behavior — all essential for invoking this correctly.

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

Parameters3/5

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

Schema description coverage is 100%, so all three parameters (projectDir, featureId, customPrompt) are already documented in the schema, which sets the baseline at 3. The description adds no further semantics such as whether projectDir must be a git repo, what form featureId takes, or how customPrompt interacts with the default feature prompt.

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

Purpose4/5

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

The description uses a specific verb and resource ('Start a Claude Code worker') and adds the execution mechanism ('in a tmux session') plus the scope ('to work on a specific feature'). An agent can tell this launches a single-feature worker rather than a monitoring or messaging tool, though it never explicitly contrasts with sibling launchers like start_continuous_worker or start_parallel_workers.

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

Usage Guidelines3/5

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

'to work on a specific feature' implies the situation for use, and 'runs autonomously' hints that no further hand-holding is required. However, there is no explicit when-not guidance and no naming of the sibling alternatives (continuous, parallel, ralph loop) that an agent must choose between when deciding how to launch work.

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