Server Quality Checklist

Tool Scores

• check_deep_researchA
  3.9/5.0
  Behavior3/5

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

  Annotations already declare readOnlyHint, openWorldHint, and idempotentHint, which cover most behavioral traits. The description adds that it returns status updates and elapsed time while in progress, which is useful context for polling. It doesn't contradict annotations. The description is adequate given the strong annotation coverage.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness4/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  The description is well-structured with a clear first sentence and a parameter/return list. It front-loads the core purpose. Some redundancy exists with the input schema (e.g., parameter defaults are repeated), but it's compact and organized for scanning. Slightly verbose with the return fields listed, but acceptable.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness4/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  The description covers the essential purpose, parameters, and return behavior. It mentions statuses and error handling. Given that an output schema exists and annotations are strong, the description is nearly complete. Minor gap: it doesn't explain the polling pattern or when the job expires, but that's acceptable for this tool.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters3/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  Schema description coverage is 0%, but the description explicitly documents both parameters: job_id as the tracking ID and include_citations with its default. It adds meaning beyond the bare schema by explaining what the job_id is and what include_citations does. This matches the baseline for tools where the description compensates for missing schema descriptions.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose5/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  The description clearly states the tool checks the status of a Deep Research job and returns the complete report once finished. It names the verb 'checks', the resource 'Deep Research job', and the specific outcome. It distinguishes itself from its sibling start_deep_research by focusing on status checking and report retrieval rather than initiation.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines4/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  The description implicitly indicates when to use this tool: after starting a job with start_deep_research, passing the job_id. It also clarifies that it returns the report upon completion, implying polling behavior. However, it doesn't explicitly state when not to use it or mention alternatives beyond the sibling, but the context makes it clear.

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

• start_deep_researchA
  4.3/5.0
  Behavior4/5

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

  Annotations already indicate non-read-only behavior (readOnlyHint=false), non-idempotency (idempotentHint=false), and open-world interaction (openWorldHint=true). The description adds that the job runs 'in the background' and returns an initial status of 'in_progress', which explains the asynchronous nature and what to expect immediately. It does not mention rate limits or failure modes, but with annotations covering the basic safety profile, this is adequate.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness4/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  The description is focused and front-loaded: the core action and asynchronous nature are in the first sentence, followed by the return values and parameter. The bullet-point structure for parameters and returns makes it scannable. It is slightly longer than absolutely necessary but every clause adds value, and there is no fluff.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness4/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  For a tool that starts a background job, the description covers the essential contract: what it does, what it returns, and how to proceed (via check_deep_research). It doesn't discuss preconditions like authentication or possible long execution times, but these are not critical for an initial call. With an output schema present, the return-value details are likely expanded there. Overall, an agent has enough to use the tool correctly.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters4/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  Schema description coverage is 0%, so the description carries the burden for parameter meaning. It defines 'prompt' as 'The comprehensive research question or topic to investigate' and marks it required. This exceeds the schema's bare 'prompt' field and gives the agent guidance on what kind of input is expected. It could add more detail like length or format, but it's sufficiently clear for correct invocation.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose5/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  The description opens with a clear action: 'Initiates a deep, multi-step web research job in the background.' It identifies both the resource (Google's Deep Research Agent) and the output (job_id), which distinguishes it from check_deep_research, the only sibling that polls status. The purpose is unambiguous and not a tautology.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines4/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  The description explicitly connects to the sibling: 'you can use to check the status of completion using check_deep_research(job_id=...)'. This tells the agent when to use this tool (to start research) and how to follow up. It doesn't state when not to use it, but given only one sibling, the workflow is clear. A minor omission is a direct 'use this when you have a new research question' phrasing, but the intent is evident.

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