run_code

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

With no annotations provided, the description carries full disclosure burden, and it does well: it states no network/host mounts, bounded output with operator-controlled truncation, 1-30 sec timeout, and that images must already exist locally (never auto-pulls). These are genuine behavioral disclosures beyond the schema about constraints, safety, and dependence on pre-existing resources.

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?

Three short, focused sentences. Each sentence earns its place: first states purpose and return type, second explains timeouts, third explains the image dependency. Zero waste, efficient and scannable.

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?

The tool is relatively simple (3 params, 2 required, output schema provided), and the description covers the key behavioral constraints (no network, no mounts, local images, timeout, truncation). It adequately covers the safety and resource-dependence properties the agent would need to know before invoking. The output schema handles return-value documentation, so the description needn't explain that further.

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 for parameter documentation. The schema covers 3 params (code, language, timeout_seconds) with enum for language. The description mentions timeout range (1-30 sec) and output truncation which contextualizes the timeout_seconds parameter. However, no additional detail is given for what the output schema returns beyond 'stdout, stderr, and status', which is stated. With 0% coverage, some description of parameter semantics is needed; it partially provides it.

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 uses a specific verb ('Run code') plus resource constraints ('without network or host mounts') and return values ('bounded stdout, stderr, and status'). It's clear what the tool does. However, it doesn't explicitly distinguish from its sibling 'sandbox_status', though the purpose is different enough that this is mostly implicit.

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 explains important constraints (no network/host mounts, timeout range 1-30 sec, requires local images) which imply when this code runner is/isn't appropriate. However, there's no explicit when-to-use vs the sibling 'sandbox_status', and no guidance on when NOT to use it or what the sandbox_status tool is for.

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