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

The readOnlyHint and openWorldHint annotations already mark this as a safe, closed-world read, so the bar is lower. The description adds real behavioral guidance beyond that: quote the text verbatim and don't paraphrase it as a promise, which signals the clause is reference material rather than a guarantee. It stays silent on how an unknown clause_id fails, which keeps it from 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?

Three short sentences: what it returns, when to call it, and how to handle the result. The purpose is front-loaded and no sentence is filler.

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 two-parameter, single-clause retrieval tool with annotations covering safety, the definition gives enough to call it correctly. The only minor gap is that it does not hint at the shape of the returned clause text or its failure mode, which would matter if the agent needs to handle a bad citation.

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 parameter formats are already documented. The description still adds provenance meaning by saying both IDs come from a citation and giving the SYN-PLAN-A / §5.2 example in context, which tells the agent where to source the values rather than just their shape.

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 opens with a specific verb and resource: 'Return the verbatim text of one clause from a synthetic plan document.' That is clearly distinct from the sibling search_plan_docs, which retrieves documents rather than a single cited clause.

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 states a concrete trigger — 'use it when the user asks why a result came back the way it did' — and describes the call pattern of passing IDs from a citation. It never names search_plan_docs as the alternative for when no citation exists, so there is no explicit exclusion.

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 and openWorldHint=false, so the safety profile is covered. The description adds genuinely useful context beyond that: the queue is worked by a person, the agent's role is reporting only, and the records deliberately exclude member/plan data. It omits ordering, volume, or pagination behavior, which is the main remaining gap.

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 short sentences with no filler. The identity and scope of the tool come first, followed by the field list and the reporting-only constraint, so the most decision-relevant information is front-loaded.

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 input parameters and no output schema, the description carries the burden of describing the return shape, and it does so by naming the four fields. What is missing is operational context: whether results are ordered, how many tickets to expect, and whether anything is filtered by default.

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 the baseline is 4; there is nothing for the description to disambiguate. The description instead spends its words on the returned fields (ticket, request_ref, reasons, created_at), which is the more valuable information for a parameterless tool.

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 specific verb and resource ('List open review tickets') plus the exact origin of the rows ('created for NEEDS_HUMAN rows') and the fields returned. It also states a negative scope boundary ('Tickets carry no member or plan data'), so an agent knows precisely what this tool surfaces without opening the schema.

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 rather than stated: the tool is for enumerating open human-review tickets, and 'the agent only reports them' signals the agent should not attempt resolution. There is no explicit when-to-use/when-not framing or comparison to the sibling tools (search_plan_docs, verify_benefits_batch, get_plan_clause), though those are clearly different domains so the risk of misrouting is low.

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 cover safety (readOnlyHint=true, openWorldHint=false), and the description adds real behavioral context beyond them: lexical rather than semantic matching, a hard cap of the top 3 clauses, the returned fields, and a privacy guarantee that member data is never returned. It does not describe empty-result behavior or how scoring ranks ties, 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?

Three short sentences, front-loaded with the mechanism, then the return contract, then the alternative-tool routing. No filler; every sentence carries distinct 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?

There is no output schema, but the description compensates by naming the exact returned fields and the top-3 cap, and it bounds scope with the no-member-data guarantee. For a two-parameter read-only search, 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 description coverage is 100% — the query parameter includes a format example and plan_id carries an enum plus its scoping meaning — so the schema already does the work. The description adds no parameter-level detail (e.g., whether plan_id affects scoring or just filtering), so the 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?

States a specific verb and resource ("keyword retrieval over the clauses of the synthetic plan documents") and immediately disambiguates the mechanism ("lexical scoring, not semantic search"), which separates it from a semantic-search sibling and from get_plan_clause. An agent can tell what this tool does without opening the schema.

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?

Explicitly routes: "Use it for general plan questions, then get_plan_clause for the full text" — a clear when-to-use plus a named alternative for the follow-up step. It lacks an explicit exclusion (e.g., when to prefer verify_benefits_batch for batch eligibility checks), so it falls short of a 5.

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 declare readOnlyHint=false and openWorldHint=false; the description goes far beyond them by disclosing the deterministic engine, the returned fact set, the NOT_VERIFIED/NEEDS_HUMAN return contract (no plan data, rule IDs in reasons, review_ticket), and that notes are stored as data and never interpreted. The audit-log storage is consistent with readOnlyHint=false, so no contradiction.

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 action and batch scope, and each following sentence contributes operational value (determinism, unverified rows, quoting rules, injection defense). It is dense and runs long as one paragraph, and the payment disclaimer is boilerplate, but nothing is redundant.

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 takes on the return-value burden and covers it, enumerating status, covered_percent, frequency, waiting period, deductible and maximum remaining, plus citations, reasons and review_ticket. An agent has everything needed to call it and interpret results correctly.

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% and every field is documented there, so the schema carries the parameter burden. The description reinforces the batch-size limit and the inert notes field but adds no format or validation detail beyond what the schema already states, and never mentions provider_id, member_id or patient_dob.

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 (verify benefits) plus the exact scope (1-25 patients, synthetic demo data). It also distinguishes itself from the follow-up tool by stating that get_plan_clause is the way to show clause text behind a citation.

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 explicit batching instructions ('call it once per batch; split longer lists into batches of 25') and an explicit alternative for a related need ('use get_plan_clause to show the text behind a citation'). It also states the negative behavior: never compute, estimate or round coverage yourself.

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