start_run

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

Annotations already indicate destructive and open-world behavior, so the description adds value by specifying the default 'read-only' permission, the persistent nature of the run, and the explicit requirement to choose a model and idempotency key. This extra context helps an agent understand the tool's default behavior and mandatory inputs beyond what annotations alone provide.

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 extremely concise: two short sentences in Chinese. It front-loads the purpose first, then adds key constraints (default read-only, explicit model/idempotency key). Every word earns its place with no filler or redundancy.

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 8 parameters, nested objects, and no output schema, the description covers the core behavior and critical requirements but leaves gaps. It does not mention prerequisites like workspace authorization (sibling `authorize_workspace`) or the `workspaceApprovalToken` parameter, nor what the return value is after starting a run. The description is adequate for a basic understanding but incomplete for a complex tool with multiple security and permission checks.

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 0%, so the description must compensate. It does address several parameters: 'workspace' (allowed local workspace), 'model' (must explicitly choose), 'idempotencyKey' (must explicitly choose), and 'permission' (default read-only). However, it leaves `task`, `timeoutMs`, `workspaceApprovalToken`, and `confirmedDangerousPermission` unexplained. The partial coverage adds meaning for some params but not all, so a baseline 3 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?

The description clearly states the tool's purpose: 'start a persistent Cursor Agent run in an allowed local workspace.' It uses a specific verb and resource, and the distinction from siblings (get_run, cancel_run, wait_run) is implicit through the action of starting a new run. It also adds scope ('allowed local workspace') and clarifies the default read-only nature, making it unambiguous.

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 implies when to use the tool (to start a run) but does not explicitly mention alternatives or when not to use it. It does not reference sibling tools like `authorize_workspace` as a prerequisite or `reply_run` for follow-up interactions. There is no guidance on choosing between this and similar tools, so the usage context is only implied.

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