Start Companion Run

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

Without annotations, the description must carry the behavioral disclosure burden. It reveals that the run is background, returns quickly with a runId, and should be monitored via other tools. This conveys asynchronous behavior and the need for polling. However, it does not mention potential side effects like process spawning, resource consumption, or backend-specific behaviors (e.g., tmux sessions for TUI backend), which are covered in the schema but not flagged as important behavioral traits.

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

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

The description is only two sentences and front-loads the primary action. The first sentence states what the tool does; the second explains the return value and directs to monitoring tools. No wasted words. It is ideal in length and structure for a tool with a rich schema.

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

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

Given the tool has 16 parameters and a rich schema, the description is concise but sufficient: it covers the core flow (start, get runId, monitor). It does not explicitly mention the distinction between new runs vs resuming/sending, but sibling names and schema cover that. It also doesn't highlight backend/security considerations, but those are in the schema. The description is complete enough for an AI agent to understand the tool's role without being verbose.

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

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

Schema description coverage is 100%, so the schema already documents all 16 parameters with detailed descriptions. The tool description itself adds no parameter-related information, but the baseline of 3 applies because the schema does the heavy lifting. The description does not need to repeat parameter details, but it also doesn't add any extra semantic context (e.g., which parameters are most commonly used together).

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

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

The description states a specific action: 'Start a background Claude Code companion run for this cwd.' This clearly identifies the verb (start), resource (companion run), and scope (current working directory). It also distinguishes itself from sibling tools by mentioning the returned runId and directing to companion_status/recent/result for monitoring, which differentiates it from companion_send or companion_resume.

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

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

The description explicitly says to use companion_status/recent/result to monitor after starting, which is clear post-use guidance. However, it does not explicitly state when not to use this tool (e.g., preferring companion_send for follow-ups or companion_resume for existing runs), so it lacks explicit exclusions. Still, the context of 'background run' and 'returns quickly' implies asynchronous starting, which is helpful.

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