Start a long-running shell command

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

Without annotations, description reveals key behaviors: returns immediately with jobId, runs over SSH, buffer limits. But it omits error handling, connection failure, and resource cleanup details.

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?

Two concise sentences: first states purpose, second gives usage context and examples. Zero wasted words.

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 no output schema, description mentions return of jobId and references related tools. It covers main use cases but lacks details on error states and output format.

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 covers 83% of parameters with descriptions; description adds no extra parameter-level meaning beyond what schema provides. Baseline score is appropriate.

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?

Title and description clearly state that exec_start starts a long-running shell command over SSH and returns immediately with a jobId. It distinguishes from sibling tools like exec by emphasizing avoidance of MCP client timeouts.

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?

Description explicitly says when to use: for long operations like git clone, pip install, etc. It directs to exec_status and exec_cancel for follow-up. However, it does not explicitly state when not to use or contrast with exec.

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