kernel_get_output

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

No annotations are provided, so the description carries the full burden. It discloses two critical behaviors: blocking/waiting semantics with timeout control, and destructive cleanup after reading ('record is cleaned up — calling again will return an error'). This is significant one-shot-consumption behavior that the agent must know. It could mention whether waiting implies any resource cost or whether partial results are returned during runs, but the provided disclosure is genuinely valuable.

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 tight paragraphs, each earning its place: purpose, behavior/timeout, and cleanup consequence. Front-loaded with the primary action in the first sentence. Zero filler or redundant restating of the tool name or 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?

For a 2-param tool with no output schema and no annotations, this description covers the critical ground: the async retrieval pattern, timeout semantics, the tag types returned, and the one-shot cleanup. It's slightly lean around edge cases (e.g., what happens on long pending runs, whether images are inlined or referenced), but for a straightforward retrieval tool the essentials are present.

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 explains msg_id implicitly as the identifier returned by kernel_execute for a pending execution, and documents the timeout parameter's role as a wait bound (or immediate return if omitted). The description adds meaningful semantics beyond the bare schema, defining both parameters' purpose and interaction.

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?

Description states a specific verb+resource ('Retrieve output for a pending execution by msg_id') and clearly delineates the scope. It names the block types returned and the cleanup behavior, distinguishing it from sibling kernel_execute which presumably returns the same blocks but executes rather than retrieves.

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 clearly explains when to use this tool (after kernel_execute, when output is pending/async), the timeout behavior (waits if running, returns immediately if omitted), and the consequence of calling again after completion (error). It doesn't explicitly contrast with alternative sibling tools by name, but the async retrieval context is strongly implied through the pending execution framing.

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