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

Annotations are neutral (non-read-only, non-idempotent, non-destructive). The description adds behavioral details: async execution, immediate job_id return, concurrency limit ('one in flight per agent' for deep_research), and estimated duration. It also clarifies the 'runs off your context' aspect. No contradictions with annotations.

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 efficient, packing type definitions, input schemas, and behavioral notes into two sentences. Each clause earns its place, though the parenthetical in 'deep_research' is lengthy. No fluff or repetition.

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 no output schema, the description explains the return (job_id) and the processes for each type, including polling instruction for deep_research. It covers concurrency limits and optional callback. It adequately prepares an agent for basic usage, though the exact output shape remains unspecified.

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 coverage is 40%, so description carries some burden. It details the 'type' enum and describes the 'inputs' object structure for each type. However, it provides minimal elaboration on 'handle', 'secret', and 'callback_url' parameters beyond their presence in the schema. The description adds value but not enough to fully compensate for the coverage gap.

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 submits an async job off the user's context and returns a job_id. It enumerates three distinct job types (fetch_bundle, delay, deep_research), each with a specific purpose. This distinguishes it from sibling tools like check_errand (polling) and research (presumably synchronous), and from other browse/archive tools.

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 delineates when to use each type: fetch_bundle for fetching URLs, delay for ping callbacks, deep_research for multi-round research. It hints at an alternative (polling check_errand for deep_research results). However, it does not explicitly state when not to use this tool (e.g., for synchronous operations) or compare to other async mechanisms.

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