start_session

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

With no annotations provided, the description carries the full burden of behavioral disclosure, but it only states the bare action. It doesn't disclose side effects (resource allocation, what happens to an existing active session), idempotency, or prerequisites for use. There is no annotation contradiction, but there is also no behavioral context beyond the verb itself.

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?

A single, tight, front-loaded sentence with zero filler — the verb and object appear immediately. It is efficient, though the brevity is partly responsible for the thin behavioral coverage penalized in other dimensions. As a structural matter, it earns its place.

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?

This tool creates a stateful resource, yet the description doesn't explain how the agent obtains a session handle from the return value (no output schema exists to fill this gap), what the timeout means for session lifetime, or how a started session is consumed by execute_code. For an agent planning a multi-step session workflow, this is insufficiently complete.

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 baseline of 3 applies even though the description adds no parameter-level detail. Both parameters (name, and timeout with its default of 900) are fully documented in the input schema, leaving nothing undocumented for the description to compensate for.

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 verb ('start') and resource ('new code interpreter session'), which is unambiguous and distinguishable from the sibling family (execute_code, stop_session, get_session, list_sessions). The word 'new' usefully signals a fresh session is created. It stops short of 5 because it doesn't explicitly contrast with execute_code, which is the most likely source of agent confusion.

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 gives no guidance on when to start a session versus calling execute_code directly, whether a session is a prerequisite for code execution, or how sessions relate to commands. An agent must infer the execution model from the name alone. No exclusions, conditions, or alternatives are mentioned.

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