Online Web Directory

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

Beyond the readOnly/idempotent annotations it discloses provenance semantics (crowd observations vs. owner_declared manifest, "never a verified fact"), how the strategy basis is derived (three independent agents vs. registry plain HTTP check with checked_at), the cost model (free daily allowance then 5 credits), and the asynchronous pending state with retry timing. These are exactly the traits annotations cannot express.

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 purpose is front-loaded in the first sentence and every later sentence carries operational content (provenance, basis, pricing, retry). It is somewhat dense with internal vocabulary (manifest_provenance, crowd_verified, registry_check) that could be structured as a short list rather than prose.

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?

With no output schema, the description still explains what comes back (posture fields, strategy, optional manifest with provenance, basis with checked_at) and covers cost and retry behavior. An agent has enough to call it and interpret the result without further documentation.

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 100%, so host, agent_token and client_class are already well documented, including the enum values. The description adds only marginal cross-cutting context — that a registered key triggers an on-the-spot check and that agent_token ties the lookup to a quota — so baseline 3 applies.

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 first sentence gives a specific verb and resource — get a site's access posture, bot challenge, JS requirement, rate limit, machine interface, plus a derived fetch strategy — which is far more concrete than the title alone. It does not explicitly name or contrast with the sibling report_observation, so sibling differentiation is left implicit.

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 timing is stated precisely: "Before fetching a website." It also tells the agent what to do in the pending case (ask again in 15 seconds). There is no explicit when-not guidance or routing to report_observation, so it stops short of full alternative coverage.

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

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

Annotations only give readOnlyHint=false, idempotentHint=true and destructiveHint=false. The description adds what those hints cannot: the operation is free/never charged, a key must be registered first, duplicates (same key, host, kind, client and time) are rejected rather than double counted, and a fact is only verified once three independent keys agree. That is substantial behavioral context beyond structured fields.

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 purpose, then prerequisites, then duplicate/verification semantics, then the return shape — a sensible information hierarchy with little filler. It is delivered as one dense paragraph, so the many distinct facts (cost, registration, dedup, verification, returns) could be broken up, but no sentence is wasted.

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 7-required-parameter, signature-authenticated write tool with no output schema, the description covers the prerequisite (registered key), the dedup rule, the verification threshold, the cost, and the exact response shape. Nothing essential for calling it correctly is missing, though reporting-side rate limits are not mentioned.

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 already 100%, so the baseline is 3. The description still adds cross-parameter meaning the schema does not state: the dedup tuple that determines rejection (key, host, kind, client, observed_at) and the rule that value must come from the fixed vocabulary for the chosen kind.

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 gives a precise verb plus resource ('report one thing you actually observed' about a fetched website) and even pins the unit of work: 'One call is one fact (kind plus value)... send one call per kind you saw.' That is far above a restatement of the name. It never references the sibling lookup_origin, so the only thing separating a 4 from a 5 is sibling differentiation.

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?

Timing and mechanics are explicit: use it 'after fetching a website,' send one call per kind, and register a key first. It gives a concrete registration path. It stops short of naming lookup_origin as the read-side alternative or stating when not to report.

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