Evident Agent Service Discovery

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

Goes beyond the readOnly/idempotent annotations by disclosing matching behavior (lexical/bilingual), that unknown auth or price is treated as unsatisfied, that results default to remote endpoints, and that pagination requires re-passing the identical query with next_cursor as after. This is meaningful operational context.

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?

Four compact sentences, front-loaded with the purpose and result-quality caveat, then matching, defaults, and pagination. Minimal waste, though the clauses are dense and telegraphic.

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?

No output schema, six params, zero schema coverage. The description covers pagination, defaults, and matching semantics but leaves several parameters (q, limit, transport) unexplained, so an agent lacks full invocation guidance despite the solid behavioral framing.

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 carry the burden. It clarifies 'after' (pass as after with identical query/filters), the transport default ('defaults to remote endpoints' maps to deployment), and the auth stance, but says nothing about q semantics, limit, or transport enum values.

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+resource: 'Discover publicly listed MCP services for a task.' Clear and distinct from the review/merchant/preparation siblings, though it never explicitly names an alternative discovery tool or contrasts itself with search_experiences/find_offerings.

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?

Implies usage ('for a task') and qualifies the result quality ('Publisher metadata, not tested recommendations'), but gives no explicit when-to-use vs alternatives, no when-not, and no prerequisites. Usage is inferrable rather than stated.

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 readOnly, idempotent, non-destructive read behavior, so the safety profile is covered. The description adds a genuinely useful semantic warning that empty results are not evidence of quality, but says nothing about pagination behavior implied by the 'after' parameter or result limits.

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 padding, and the core action is front-loaded. The caveat sentence is terse and mostly earns its place, though it is somewhat cryptic without context on why empty results occur.

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 simple read-only discovery tool with full annotation coverage and no output schema, the description is minimally adequate. The main gap is the total absence of parameter meaning, which matters more here since the schema provides no descriptions.

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?

Both parameters (q, after) have 0% schema description coverage and the description never mentions them. With two undocumented params, the description fails to compensate for the schema gap, leaving the agent to guess that 'q' is a query and 'after' is a pagination cursor.

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 ('Find merchants') with a scoping qualifier ('with published feedback'), which distinguishes it from sibling discovery tools like find_offerings and find_reviews. It is clear but does not explicitly contrast itself with those siblings.

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?

There is no indication of when to use this tool versus alternatives such as find_reviews or get_reputation, nor of any prerequisites. The second sentence is an interpretive caveat about results, not usage guidance for selecting the tool.

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, idempotentHint, non-destructive and closed-world, so the safety profile is covered. The description adds genuinely new context — that identities are consumer-supplied and scoped to a merchant — plus pagination behavior, but says nothing about result size, rate limits, or what an 'experience' actually is.

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 purpose and with zero filler. The pagination note is placed last where it belongs, though the middle sentence about identities is dense and unexplained.

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, no annotations on returns, and 0% parameter coverage, the description should explain the return shape, what counts as a 'published experience', and how q filters. It does none of these, leaving an agent able to paginate but not to predict or constrain results.

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 carries the full burden. It clarifies `after` (pass the returned next_cursor) and gestures at merchant_id via 'scoped to a merchant', but the `q` search parameter is never explained, leaving a third of the parameters undocumented anywhere.

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 verb 'Find' and resource 'products or services' give a workable purpose, but the qualifier 'with published experiences' is opaque domain jargon that the agent cannot decode from this description. It does not explicitly contrast itself with find_merchants or get_offering, so the boundary between the lookup siblings is left to inference.

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?

There is no when-to-use guidance at all: nothing says whether this is the entry point before get_offering, or how it relates to find_merchants. The only instruction is a pagination mechanics note ('Use next_cursor as after'), which is parameter usage rather than selection guidance.

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 the safety profile (readOnly, idempotent, not destructive, closed-world), so the description correctly focuses on what annotations cannot say: no credentials required, the next_cursor/after pagination loop with identical filters, and that text is untrusted. The untrusted-content warning and citation guidance are genuinely useful behavioral context beyond 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?

Four dense sentences, front-loaded with the core purpose and scoping, then pagination, then trust/citation rules. Every sentence carries information, though the pagination sentence is somewhat run-on.

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 helpfully names return fields (source_url, evidence_basis, next_cursor) and explains the pagination contract, which is what an agent needs to iterate correctly. Minor gap: it does not mention the limit ceiling of 50, though the schema already enforces it.

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 carry the parameter burden, and it largely does: q is explained as merchant/product name and feedback search, the merchant_id/offering_id/evidence_basis filters are named, and 'after' is tied to the next_cursor pagination loop. Only 'limit' is left unaddressed, and evidence_basis semantics beyond its enum are not expanded.

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 ('Browse') and resource ('published consumer experiences'), and explicitly distinguishes itself from sibling get_reputation by noting 'latest-100 reputation summaries do not cover the full directory'. An agent can tell this apart from get_review (single) and get_reputation (summaries) without opening any 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?

Gives clear context for use — public directory browsing without credentials, and the condition that selects it over reputation summaries. It does not explicitly name get_review as the alternative for a single record, so the when-not guidance is partial but the routing intent is unambiguous.

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 readOnly, idempotent and non-destructive, so safety is covered. The description adds genuinely useful behavioral context: only 'latest published' reviews are returned, ratings are 'evidence-separated', and excerpts are explicitly not consensus or recommendations, which shapes how an agent should interpret and cite output.

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 tight sentences with no filler; the return contents are front-loaded and the interpretive caveat follows. Nothing is wasted, though the second sentence mixes usage guidance with output caveats.

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 does convey what comes back (reviews, ratings, brief), but omits volume/pagination, error behavior for a missing or unpublished offering, and how offering_id is sourced. Adequate but with clear gaps for a retrieval tool.

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% and the description never explains the single required offering_id — its format, or that it is obtained from find_offerings. With one undocumented parameter, the description fails to compensate for the schema gap.

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 uses a specific verb ('Read') and resource ('one product/service') and enumerates the returned payload (reviews, evidence-separated ratings, experience brief). It is clear what the tool does, though it does not explicitly distinguish itself from siblings like get_reputation or find_offerings.

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?

There is no explicit when-to-use guidance or mention of alternatives such as find_offerings for discovery versus get_offering for retrieval by id. The citation directive hints at usage but does not route the agent between sibling tools.

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, idempotentHint, destructiveHint=false and a closed-world scope, so the safety profile is covered. The description adds genuinely useful behavioral context by restricting the data exposed to just draft ID and status, excluding other drafts and raw files.

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, front-loaded with the action and the returned scope, with no filler. Every clause earns its place by either defining the output or bounding access.

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 helpfully enumerates what is returned (draft ID and status), which compensates well. The one gap is that it does not explain how the review:prepare token is supplied, given the schema takes no parameters.

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?

There are zero parameters, so the baseline is 4. The description adds no parameter detail because none exist; the only implicit input, the review:prepare token, is not parameterized in the schema.

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 (read) and a narrowly scoped resource (the draft ID and status tied to a review:prepare token). The scope phrases 'only' and 'this review:prepare token' clearly separate it from read-heavy sibling tools, though no sibling is named directly.

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: call it to inspect the draft associated with a prepare token, and the 'No access to other drafts or raw files' clause sets a negative boundary. However, it never states explicitly when to prefer this over siblings like get_offering or submit_review, nor any prerequisite ordering (e.g., after prepare_review).

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, idempotentHint=true, destructiveHint=false, and openWorldHint=false, so safety is covered. The description adds a real behavioral note — "Identity is consumer-supplied" — and a result-scoping constraint ("with published experiences"), which is context beyond the annotations. It still says nothing about pagination or response shape, so this is modest added value.

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, and the core action is front-loaded. The trailing clause about identity and feedback is compact but slightly cryptic, costing a little clarity rather than length.

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 annotations covering the safety profile and no output schema to explain, the description only needs to cover scope and usage. It establishes that results are seller offerings with published experiences, but leaves the get_product vs listing-offerings ambiguity and the meaning of "identity is consumer-supplied" unexplained. Adequate but with clear gaps for a 1-param read tool.

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% and the single required parameter has no description. The description does reference the product ID, implicitly confirming that the param selects the product whose offerings are returned, but it adds no format, ID scheme, or validation detail. Marginal compensation for the coverage gap.

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 verb ("List") and a resource ("seller-specific offerings for a product ID"), but the resource conflicts with the tool name get_product, which suggests retrieving a product rather than listing offerings. The qualifier "with published experiences" is vague and does not distinguish this from siblings like find_offerings or get_offering. The purpose is inferable but not crisp.

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?

"Keep seller service feedback separate" hints at scope boundaries but names no alternative tool and states no condition for choosing this over find_offerings/get_offering. There is no explicit when-to-use or when-not-to-use guidance. Usage has to be inferred entirely.

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 readOnly/idempotent/non-destructive, but the description adds real behavioral value beyond them: 'Transaction verification is always false' discloses a data characteristic the agent would otherwise misread, and the untrusted-data warning is an explicit safety directive. It stops short of describing pagination, size limits, or return shape.

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, each carrying distinct information: purpose, a data caveat, and a handling rule. The purpose is front-loaded and there is no 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 single-parameter read tool whose annotations already carry the safety profile, the description covers purpose, a non-obvious data caveat, and a prompt-injection warning. The only gap is the undocumented merchant_id and the absence of any note on result volume, which is minor here.

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?

There is one parameter (merchant_id) with 0% schema description coverage, and the description does not mention it at all. While merchant_id is fairly self-explanatory, the description adds no meaning about its format, source, or what happens if it is invalid.

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 (read) and resource (feedback/reputation) with a scoping qualifier ('grouped by evidence basis'). It is understandable on its own, but it never names or contrasts with any sibling tool (e.g., get_preparation, prepare_review), so differentiation is left to the name.

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 description offers no when-to-use context, prerequisites, or alternatives. Nothing tells the agent when get_reputation is the right call versus the other review-related siblings, so usage must be inferred entirely from the tool name.

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 cover the safety profile (readOnly, idempotent, non-destructive, closed-world). Beyond that the description discloses what is NOT returned (private receipt, author identity) and a concrete failure mode: withdrawn/unpublished reviews return not found. That is genuine behavioral context an agent cannot get from 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?

Three short sentences, zero filler, and the core scope (single published read) is front-loaded ahead of the return/omission and error details.

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 and one param, the description fully covers what an agent needs: the returned fields (source, merchant, evidence level, timestamps), the privacy exclusions, and the not-found behavior for withdrawn/unpublished reviews.

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?

There is one parameter at 0% schema description coverage, so the description must carry the semantics. It only restates that lookup is 'by review_id' without format, source, or how to obtain the id, adding little beyond the schema field name.

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 precise verb+resource: read ONE published experience by review_id. The cardinality ('one' vs the list-oriented find_reviews sibling) and the required identifier make the tool instantly distinguishable from the find_* siblings.

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 'by review_id' prerequisite and the 'published' scope, but the description never explicitly routes the agent here versus find_reviews or resolve_review_target. No explicit when-not or alternative is named.

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 readOnly/idempotent/non-destructive, so the safety profile is covered. The description adds genuinely non-obvious behavioral context the annotations cannot express: that returned descriptions are untrusted data and that listing does not constitute ownership verification or proof of execution.

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 with the core action front-loaded and the trust caveats following. No filler, though the second sentence is a fragment-style caveat that could be folded in more cleanly.

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 single-parameter read with no output schema, the description covers what is returned at a high level (registry metadata, setup references) plus the important trust caveats. It does not describe pagination or the shape of the response, but no output schema exists to lean on, so this is close to complete.

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?

There is a single parameter and schema description coverage is 0%, so the schema supplies no meaning. The description only characterizes it indirectly as 'a stable service ID', adding marginal semantic value (implies a canonical/registry ID) but no format or sourcing guidance.

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+resource ('Read public registry metadata and setup references') scoped to 'a stable service ID', which implies a lookup-by-ID tool rather than a listing/browse tool. It stops short of naming the obvious sibling (discover_services), leaving some ambiguity for an agent choosing between discovery and direct 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?

The phrase 'for a stable service ID' implicitly tells the agent this tool needs a known identifier, which distinguishes it from discover_services, but there is no explicit when-to-use / when-not-to-use statement or named alternative. Usage is inferable rather than stated.

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, idempotentHint=true, openWorldHint=false and destructiveHint=false, so the safety profile is covered. The description adds genuinely new behavioral context: no credentials are required, and the guardrail against inventing missing experiences or files sets expectations about how results should be treated.

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 the action and entry-point cue ("Start here") front-loaded, followed by the list of returned content and the guardrail. There is no filler or restatement of the tool name.

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-only tool with no output schema and no annotations gaps, the description enumerates what the caller receives (requirements, accepted evidence types, minimum inputs, confirmation steps). Nothing needed 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?

This tool takes zero parameters, so the baseline is 4 under the scoring rules. The description correctly adds nothing about parameter syntax because there is none, and the schema is fully closed with additionalProperties=false.

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 pairs a specific verb (read) with a specific resource (current submission requirements) and enumerates the content returned: accepted private evidence, minimum inputs, and consumer confirmation steps. It does not explicitly contrast itself with sibling tools like get_preparation or upload_evidence, so it stops short of full 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?

"Start here" gives a clear sequencing cue that this is the entry-point call before prepare_review/submit_review, which is meaningful context for an agent picking among eleven siblings. However, no alternative is named and no when-not condition is stated, so it is context rather than explicit 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?

Annotations already declare readOnlyHint, idempotentHint and destructiveHint=false, so the safety profile is covered. The description adds genuinely new behavioral context beyond that: it is a dry-run that persists nothing and returns 'only consumer-provided facts' rather than derived judgments. That is exactly the kind of extra context this dimension rewards.

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?

Four short sentences with the core action front-loaded and no filler. The closing sentence 'Return only consumer-provided facts' is terse to the point of being slightly cryptic, but nothing 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?

With no output schema, the description must carry the return-value burden, and 'Return only consumer-provided facts' only partially does so — it doesn't say what the check result looks like or what a failure signals. Combined with 9 undocumented nested parameters, the definition is adequate to start a call but leaves meaningful gaps.

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?

Nine parameters, several nested (file with mime/size constraints, target with 18 identity fields), at 0% schema description coverage. The description only alludes to 'proposal fields and file metadata' generically and documents no parameter, its format, or which of the many target identity fields matter. It fails to compensate for the coverage gap.

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: 'Check proposal fields and file metadata before authorization/upload,' and immediately clarifies the scope is validation-only ('No files or reviews are saved'). This distinguishes it from submit_review/upload_evidence siblings, though it never names those siblings explicitly, so an agent must infer the routing.

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 clear timing context ('before authorization/upload') and explicit exclusions of what it will not validate (credentials, receipts, duplicate evidence, approval), which usefully bounds expectations. However, it names no alternative tool and gives no condition for choosing this over prepare_review or submit_review.

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 the write/destructive profile, and the description adds real value on top: the required review:prepare token (an auth prerequisite), the immutability of the created proposal, the fact that nothing is confirmed or published, and the returned confirmation URL. The one gap is that destructiveHint=true is never explained — the description never says what state is overwritten or consumed.

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?

Four sentences, no filler, with the core action and the token requirement front-loaded before the conditional field rules. The dense conditional clauses are compact but readable; nothing is padding.

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 9-parameter tool with nested objects and no output schema, the description covers the return value (confirmation URL) and the hardest identity-branching logic, which is genuinely useful. It still leaves the required evidence_id, feedback, and incentivized parameters and the whole context block unexplained, which is a meaningful gap for a preparation call.

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% across 9 parameters (many nested), so the description must compensate and it only partly does: it explains the target.merchant_id/offering_id branching, merchant_name, merchant_website, offering reuse, and that ratings are optional and must not be guessed. evidence_id, feedback, incentivized, and the entire context object are left undocumented in both schema and description.

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 ("Prepare one immutable proposal under a review:prepare token") and implicitly separates itself from the publishing step with "This does not confirm or publish it." It does not, however, differentiate itself from the similarly named sibling preflight_review or get_preparation, which an agent must disambiguate on its own.

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 context is only implied: "does not confirm or publish" hints that submit_review comes later, but no explicit when-to-use statement or sibling routing is given. The conditional target-field rules (merchant_id alone, offering_id alone) are field-filling instructions rather than selection guidance, so the agent gets partial, mostly inferential direction.

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 readOnly, idempotent, non-destructive, so the safety profile is covered. The description adds genuinely new behavioral facts: it performs no authentication, booking, or payment, and importantly creates no service-use record or review. That last point is a side effect an agent could not infer from annotations alone.

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?

Four short sentences, front-loaded with the core action before the negative clauses, so the key fact is read first. The fourth sentence's 'no service-use record' is partially redundant with the second sentence's 'does not ... book or pay', but the wording is otherwise tight.

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 single-parameter read-only tool with annotations and no output schema, the behavioral picture is reasonably complete and return values need not be explained. The gap is the missing parameter guidance and the absence of any pointer to which sibling produces a valid service_id.

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% and the description never mentions service_id at all. With a required, undocumented parameter the description should compensate by saying what the identifier refers to or where to obtain it (e.g. from discover_services or get_service); it does not.

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 states the specific resource retrieved (connection instructions) and sharply delimits it with a list of things it does not do (execute, authenticate, book, pay). This distinguishes it from siblings like get_service, get_preparation, and preflight_review, though it never states outright that it is keyed to a service and yields handoff details rather than a service listing.

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?

"Connect from your own client after confirming task fit, permissions and costs" implies the intended workflow and the prerequisite checks, but there is no explicit when-to-use-this-versus-alternatives routing (e.g. versus get_preparation or get_service). Usage is inferred rather than stated.

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 destructiveHint=true and non-idempotent behavior; the description's 'publish or replace' is consistent with that and clarifies that an existing response is overwritten. It also adds genuinely new context beyond annotations: an authorization prerequisite (approved merchant:reply token) and a business-scoping constraint.

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?

A single front-loaded sentence with no wasted words, leading with the action before the qualifying conditions. It is dense with domain jargon ('merchant:reply token') but not padded.

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 2-parameter mutation with no output schema, the description covers purpose, authorization prerequisite, and scope, which is a reasonable floor. It still omits what the body parameter should contain and the concrete effect of replacing an existing response, leaving an agent to guess at the destructive behavior's details.

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% for two required parameters (review_id, body), so the description must carry the meaning and largely fails to. It only implies that 'body' is the merchant response text via 'merchant response'; review_id is never explained, nor are format or length expectations.

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 specific verb pair (publish/replace) and the resource (merchant response), and clarifies it is a reply to a review rather than a submission or preparation step. It does not explicitly name siblings like submit_review or prepare_review, so differentiation relies on inference from the 'merchant:reply token' phrase.

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 phrase 'under an approved merchant:reply token' implies a prerequisite step (a token must already be approved, likely via prepare_review) and bounds usage to that business. However, it never states when to choose this tool over siblings such as submit_review or prepare_review, leaving usage only implied.

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?

Goes well beyond the readOnly/idempotent/non-destructive annotations: new_target IDs are reserved deterministically but persisted only upon a review, names return candidates without fuzzy auto-merge, there is no transaction or merchant verification, and consumers must check seller and variant. These are exactly the behavioral caveats an agent needs before trusting a resolved ID.

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?

Dense but front-loaded: purpose first, then identifier precedence, then behavioral caveats. A few sentences are run-on (the 'Otherwise use ...' clause packs six parameters into one list), but nothing is padding.

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 18 parameters, no output schema, and no parameter descriptions, the description covers the resolution logic and the persistence caveat but is thin on the return shape ('Returned history paths support immediate reading' is opaque) and on which of the many optional identifiers are mutually exclusive or combinable. Adequate, with clear gaps for a tool of this complexity.

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% across 18 parameters, so the description must carry the load. It explains the meaning and combination of the core identifier chain (identity_source, provider_id, location_id, item_id/item_kind/item_version, store_url, merchant_id, offering_id), but leaves gtin, listing_id, marketplace, channel_url, product_url, product_name, merchant_name, seller_id and variant without any stated semantics or combination rules.

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 ('Resolve') and resource set (providers, locations, items, services) with an outcome ('stable identities'). It is distinguishable from the find_*/get_* siblings, though it never explicitly contrasts itself with find_merchants or find_offerings, which also handle merchants/offerings.

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 a clear precedence order for identifier paths (merchant_id/offering_id first, then identity_source+provider_id chain, then ecommerce identifiers), which is operational guidance. However it never says when to reach for this tool versus searching with find_merchants/find_offerings, and no prerequisites or exclusions are stated.

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, idempotentHint, destructiveHint=false, so the safety profile is covered. The description adds genuinely new behavioral context: URLs are not fetched (important for an agent expecting fetching to happen), and names/URLs are candidate clues rather than verified identity — a reliability caveat the 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?

Three tight sentences with no filler; the invocation form is front-loaded and the caveats are ordered after it. Slightly compressed to the point of ambiguity, but nothing 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?

There is no output schema, and while the description sketches returns ('original reviews, matching context, coverage and next action'), it is thin for a tool whose only structured input is a twenty-field nested identifier object with zero field descriptions. The complexity of the target object is simply unaddressed.

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 carries the full burden. It does explain q (name or public provider URL) and gestures at 'target identifiers,' which is useful, but limit is never mentioned and the nested target object's twenty identifier fields (gtin, seller_id, listing_id, store_url, etc.) are entirely unexplained in both schema and description.

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 verb phrase is 'start a public lookup,' which is soft, and the resource is left implicit — the name says 'experiences' but the description says it returns 'original reviews.' It lists the inputs (q name/URL or target identifiers) and the return shape, but never distinguishes itself from siblings like find_reviews or resolve_review_target, so an agent can't confidently pick it without reading schemas.

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 tells you how to invoke it (q vs target identifiers) but not when to use it versus find_reviews, find_merchants, find_offerings, or resolve_review_target. There is no stated precondition, no exclusion, and no routing guidance among a large sibling set.

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 declare a destructive, non-idempotent write; the description adds real context beyond them — that the outcome may be immediate publication or human-review queuing, that a distinct submit token is required, and a warning not to conflate publication with transaction verification. It stops short of covering irreversibility or duplicate-submission behavior that the destructive/non-idempotent hints imply.

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 tight sentences, front-loaded with the action and its required input, then the branching outcome, then the caveat. Each sentence carries distinct information; no 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 single-parameter tool with no output schema, the description supplies the outcome variability, the status-reading instruction, and the semantic caveat, which is enough for correct invocation. Only the draft_id semantics and duplicate-submission risk are left unaddressed.

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?

Only one parameter (draft_id) with 0% schema description coverage, so the description must compensate. It references the 'consumer-confirmed draft' and a separate review:submit token, but never clarifies what draft_id refers to or where the token comes from — and the token isn't even a parameter here.

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 ('Submit') and resource ('consumer-confirmed draft'), which cleanly separates it from the sibling prepare_review that produces the draft. The jargon 'review:submit token' and 'self-reported review' are domain-specific but do identify the operation.

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 implies the prerequisite (a draft plus its separate review:submit token) and tells the agent to read the returned status, which is useful. However, it never explicitly states when to use this instead of prepare_review or reply_to_review, leaving routing to inference.

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 flag the write/destructive/non-idempotent profile, and the description adds genuinely new operational context: size ceiling, accepted MIME types, base64 transport, auth scope, token expiry window, and a consumer-consent policy. It does not explain what happens on success or why the operation is destructive, keeping 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, zero filler, with the most decision-critical facts (what it accepts, size, dependency) front-loaded. 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 single-parameter upload tool with annotations covering the safety profile and no output schema, the description covers prerequisites, limits, consent policy, and encoding. The only gap is post-upload behavior (return value, duplicate handling despite non-idempotentHint), which is minor.

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% for the single file_base64 parameter, so the description must compensate — and it does, specifying encoding (base64), accepted formats, and the 5 MiB cap. It does not describe the exact base64 payload shape or whether a filename accompanies it, so not a 5.

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+resource (upload evidence) with tight qualifiers: private, consumer-authorized, PDF/PNG/JPEG, max 5 MiB, base64. This clearly distinguishes it from siblings like submit_review or prepare_review without opening any 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?

The description states a hard prerequisite ('Requires review:prepare via OAuth or a one-hour manual Bearer token'), which tells the agent this tool belongs to the review-preparation flow. It does not name an alternative tool or an explicit when-not-to-use case, 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.