Ocolta Review Summaries

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

Annotations already declare readOnlyHint=true, destructiveHint=false, and openWorldHint=false, so the safety profile is largely covered. The description goes beyond that by enumerating the returned facts (label, synthetic-fixture flag, unused review count, report-summary access) and explicitly naming what is withheld ('Does not expose payment details or account identifiers'), a clear data-boundary disclosure.

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?

Three compact sentences, front-loaded with the identity of the resource, then the return contents, then the privacy boundary. Every sentence carries information; no filler or restatement of the title.

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 zero-param, read-only tool with an output schema, the description covers purpose, scope, return contents, and data exclusions. The only unaddressed detail is behavior when no connected purchase exists (e.g., error vs. empty result), which is a minor gap given the output schema already documents the success shape.

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?

The tool takes zero parameters, so per the rubric the baseline is 4. There is nothing parameter-level for the description to compensate for, and the schema with additionalProperties=false is unambiguous.

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?

States a specific verb and resource: 'Identifies the existing Ocolta purchase connected to this conversation.' An agent can tell this is a lookup of the purchase bound to the current conversation. It does not, however, distinguish itself from the siblings get_review_summary and list_recent_reviews, even though it mentions reviews and report-summary access.

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?

Usage is implied by the phrase 'connected to this conversation', which tells the agent the tool is scoped to the current conversation. There is no explicit when-to-use, when-not-to-use, or comparison against the review-oriented siblings, so the agent must infer the routing.

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?

Beyond annotations (which already declare readOnly/openWorld=false/destructive=false), it discloses several traits the agent could not otherwise know: no credit is spent, a fixed disclaimer is attached, and the response deliberately omits source documents, personal fields and free-text excerpts. It also bounds interpretation by stating the result does not prove fraud, authenticity, intent, identity or AI generation.

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?

Three sentences, zero preamble: what it returns, what it withholds/costs, and how to interpret it. Each sentence earns its place, though the non-goal enumeration ('fraud, authenticity, intent, identity or AI generation') is slightly long for its informational payload.

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 an output schema present, return-value structure need not be spelled out, and annotations already cover the safety profile. The description adds the remaining essentials for a fraud-adjacent read tool: cost, data minimization, and explicit interpretation limits, so nothing an agent needs to call it correctly is missing.

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 0%, so the description must carry the load, and it only partially does: it clarifies that the ID must reference a single recoverable review tied to the connected purchase, which scopes the value. It adds nothing about the required ent_<hex32> format or where the ID is obtained, leaving the regex pattern in the schema to do that work.

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?

States a specific verb (Returns) plus the exact resource (categorical status, priority, confidence, fraud/AI assessments, indicator and limitation counts) for one recoverable review from the connected purchase. The singular 'one ... review' scoping implicitly separates it from the sibling list_recent_reviews, and the 'from the connected purchase' constraint separates it from a free-standing review lookup.

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 decision-relevant context (it consumes no credit, and output is a signal for human follow-up rather than proof), which helps an agent judge whether to call it. However, it never states when to prefer it over list_recent_reviews or get_connected_purchase, nor prerequisites such as where a valid review_id comes from. Usage is implied rather than prescribed.

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 already declare readOnlyHint=true, destructiveHint=false and openWorldHint=false, so safety is covered; the description adds substantive behavior beyond them: a hard cap of ten results, a 24-hour recovery window, most-recent-first ordering, and the fact that calling it is free (no credit spent). This is real operational context rather than a restatement of the 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?

Two tightly written sentences with zero filler: the scoping constraints (ten max, connected purchase, 24-hour window, ordering) come first, and the negative clarification about not starting a review or spending a credit lands last. Every clause earns its place.

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 zero-parameter read tool with an output schema, the description carries what the agent needs and need not explain return values. The only soft spot is that 'the connected purchase' is referenced without definition, though the sibling get_connected_purchase implies how that context is established.

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?

The tool takes zero parameters, which sets the baseline at 4; there is nothing to document and the description correctly makes no parameter claims. Schema description coverage is reported at 100% and the empty object schema is unambiguous.

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?

Specific verb (lists) plus a precisely scoped resource (review IDs of up to ten completed Deep Reviews from the connected purchase), including the 24-hour recovery window and most-recent-first ordering. It does not name the sibling tools (get_connected_purchase, get_review_summary), so it falls short of explicit sibling differentiation, but the negative clause 'Does not start a review or spend a credit' fences off adjacent actions.

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?

Clear context for use: only reviews still inside the 24-hour post-completion recovery period are returned, and the description explicitly rules out starting a review or spending a credit. It stops short of naming what to call instead (e.g., get_review_summary) for reviews outside that window, so no explicit alternative is offered.

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