search

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

With no annotations at all, the description carries the full burden and does so: it discloses the two distinct return shapes, that web-provider results carry a `session` token for follow-up turns, that chatgpt_web performs a full chat turn with web search on by default, that hosted HTTP mode rejects `file` inputs due to filesystem access, and that providers are consulted in a priority order unless forced. These are non-obvious traits an agent could not infer from the schema.

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?

Front-loaded with the core purpose, then layered detail about result shapes, provider-specific behaviour, and where to look for the long tail. It is dense rather than verbose — every sentence carries information, though the parenthetical shorthand ('Niche knobs', 'structured extract…') is slightly cryptic on first read.

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?

Ten parameters, no annotations, and no output schema — the description compensates by explaining return shapes, session continuation, and the mode/model knobs it deliberately does not enumerate, delegating that to `usage`. Nothing an agent needs in order to call it correctly or interpret the response is left unstated.

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 100%, so the baseline is 3; the description still adds cross-parameter meaning the schema lacks — that a follow-up message goes in `query` when `session` is set, that `file` implicitly targets grok_web when no provider is forced, and that `topic` steers backend selection rather than just labelling a niche. It does repeat some schema text (recency, domains, max_results) without enrichment.

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?

Names a specific verb and resource ('Web search') plus the scope ('across quota-aware providers'), and splits the tool's behaviour into two concrete result classes: ranked title/url/snippet results from API providers versus synthesized answers with sources from web providers. An agent can tell what it will get back without opening the schema, and the description points to `usage` for anything outside this tool's core remit.

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?

Gives concrete conditional guidance: use `mode="chat"` on chatgpt_web to answer from the model alone, attach `file` only in local stdio mode, and call `usage(provider=…)` for provider-specific extras like scholar/patents/places. What is missing is routing against the siblings — nothing says when to prefer `deep_research` or `browser` over this tool, and there are no exclusion conditions.

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