deep_research

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

With no annotations supplied, the description carries the full burden and does substantial work: it discloses the internal loop, an upper bound via max_pages, the shape of the return value (numbered sources with supporting passages), and a prompt-injection warning that passages are data, not instructions. It does not mention cost, latency, or failure behavior when pages cannot be fetched, which keeps it below a 5.

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 pipeline and output come first, followed by the usage trigger and the injection caveat, so the important information is front-loaded. It is slightly run-on in the opening sentence and tucks the only parameter hints at the very end, but almost every clause carries information.

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 five-parameter, no-annotation, no-output-schema research tool the description covers the workflow, the return format, and the injection risk well. The remaining gap is the unexplained depth and breadth knobs, which materially change how many searches and sources the agent should expect.

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 define all five parameters, and it only partially does: it explains max_pages ("reads up to max_pages pages") and queries ("your own search angles, optional"). The `depth` and `breadth` parameters — both defaulted integers with no schema description — are never explained, leaving an agent to guess their effect on the pipeline.

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 names a concrete multi-step operation (search several angles, read pages, follow links, re-search, return numbered sources with answering passages) rather than restating the name. It is clearly distinguishable from siblings like search, ask, or agent because the iterative read-and-follow pipeline and the cited-passage output are spelled out.

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?

It gives an explicit trigger: use this whenever the user asks for "deep research." That is a clear selection rule, but it does not say when NOT to use it — for example how it differs from the sibling `agent` or `ask` when the request is broad but not literally labeled deep research, or when to prefer plain `search`.

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