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

Annotations already declare readOnlyHint, idempotentHint, destructiveHint=false and openWorldHint, so the safety profile is covered. The description contributes one genuinely extra behavioral fact — that no latest-version guessing occurs and only a provider-marked default is used — which is real resolution semantics not in the annotations. It stops there: no mention of network dependency, error behavior when a default is absent, or supported ecosystems.

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 ahead of the caveat. It is terse rather than bloated; the only cost is that brevity here shades into under-specification.

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?

An output schema exists, so return values need not be explained, and annotations carry the safety profile. But for a tool keyed on package name plus ecosystem, the description says nothing about the ecosystem parameter or its npm default, and gives no signal about what happens when no provider-marked default exists — the exact failure mode its own caveat implies.

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 both parameters: the description never mentions `name` or `system`, nor that `system` defaults to npm, nor which ecosystems are valid. With zero coverage the description is expected to compensate, and it does not — the only hint is the generic phrase "version metadata."

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?

"Get deps.dev version metadata" names a verb and a resource, but "version metadata" is vague — it doesn't say whether this returns package info, resolved versions, or dependency data. The second sentence clarifies the resolution policy rather than the purpose. Siblings (osv_*, nvd_cve, kev_check, epss) are vulnerability tools, so the resource is distinguishable, but the description never states that distinction itself.

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?

No when-to-use, when-not-to-use, or alternative routing is given — nothing tells the agent why it would pick deps.dev metadata over the OSV/NVD/KEV siblings. The note about only accepting a provider-marked default hints at a resolution use case, but that is an implicit inference rather than stated 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 declare readOnly, idempotent, openWorld and non-destructive, so the safety bar is low; the description still adds real value by disclosing the one-page-per-call pagination model and by warning that scores are exploitation-probability estimates rather than environment-specific risk. It omits rate limits, result ordering, and behavior on invalid CVE IDs.

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, no filler, with the operational constraint (one page, 100 max) front-loaded and the interpretive caveat second. Every sentence 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?

An output schema exists, so return values needn't be explained, and annotations cover the safety profile. The description supplies the paging model and the score-interpretation caveat, which is most of what an agent needs; only pagination mechanics and alternatives are thin.

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 has to carry parameter meaning and it only partially does: "up to 100 CVEs" maps to cve_ids/maxItems and limit's cap, but offset is never mentioned and pagination semantics are left implicit. The CVE-ID format and bounds come entirely from 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?

Specific verb ("Fetch") plus resource ("FIRST EPSS page") and scope ("up to 100 CVEs") let an agent immediately tell this is the exploitation-probability scoring source, distinct from the OSV/NVD/KEV siblings that return advisory or catalog data. It stops short of explicitly naming a sibling it should be preferred over, but the resource is unambiguous.

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 or when-not-to-use guidance and no routing to alternatives such as kev_check or nvd_cve. "Fetch one ... page" implies the result is paged but never says how to continue to the next page or when to prefer a single-CVE lookup, so the agent must infer usage.

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, openWorld, and non-destructive, so safety is covered. The description adds genuinely useful behavioral context beyond that: the catalog is 'freshly fetched' (a live network lookup with current data) and an empty result is explicitly not proof of no exploitation. It does not describe return shape, but an output schema exists to cover that.

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 core action front-loaded and the interpretive caveat second. No filler, no repetition of schema details beyond the count limit.

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 low-complexity, single-parameter lookup with an output schema and full safety annotations, the definition covers purpose, data freshness, and the key negative-result caveat. The one meaningful omission is guidance on when to reach for this tool versus the other vulnerability-data siblings.

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 required parameter with 0% schema description coverage, so the description could compensate more. It confirms the parameter is a list of CVEs capped at 100, but that cap is already enforced by schema maxItems and the CVE-ID format is only expressed in the schema pattern, not in prose.

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 (check) and a specific resource (CISA KEV catalog) with a scope limit (up to 100 CVEs). Naming the CISA KEV catalog uniquely distinguishes it from sibling tools like nvd_cve, osv_query, and epss, which cover different vulnerability data sources.

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 says what the tool does but never says when to use it instead of nvd_cve, osv_vuln, or epss, nor when not to use it. 'Absence is not proof of no exploitation' is a result caveat, not usage guidance, so the agent must infer routing from the name alone.

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, and open-world, so the safety profile is covered. The description adds genuine operational context beyond annotations — the environment-read API key and the fact that metric sources are kept distinct rather than merged — but says nothing about behavior without a key (rate limits, error mode) or pagination/response size, which matters for an open-world external call.

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 zero filler, and the core action is front-loaded before the credential note. Every clause 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?

An output schema exists, so return-value documentation is not required, and the credential note covers the main setup surprise. What remains missing is the routing context against four sibling vulnerability tools, which is the largest gap for a single-purpose lookup.

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 only format guidance is the machine-readable pattern ^[cC][vV][eE]-[0-9]{4}-[0-9]{4,}$ and maxLength 64. The description never mentions cve_id at all, so it neither confirms the CVE-YYYY-NNNN format nor explains the looser 4+ digit sequence; the parameter is largely self-evident, keeping this at baseline rather than lower.

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 ("Get an NVD CVE") and adds that metric-source attribution is preserved, which hints at a distinctive return shape versus a plain lookup. It does not, however, distinguish itself from siblings like osv_query, osv_vuln, or kev_check — the agent must infer the NVD-vs-OSV-vs-KEV split from the name alone.

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 statement of when to use this tool versus osv_vuln, osv_query, kev_check, or epss, all of which surface vulnerability data. The only usage-adjacent content is the note that NVD_API_KEY comes from the environment, which is a precondition rather than a routing rule.

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 context: results are paginated per package and each package's page_token must be followed independently — a non-obvious trait worth stating. It says nothing about rate limits, partial failures, or how results are keyed back to packages, though the output schema may cover the latter.

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 item shape and capped by the batch limit. Terse to the point of near-telegraphic ambiguity, 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?

An output schema exists, so return-value explanation is unnecessary. For a single-parameter, read-only batch tool the description covers shape, cap, and pagination follow-up, but leaves purpose framing and comparison with osv_query implicit.

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 compensate. It does enumerate the per-package fields (name, version, ecosystem?, page_token?) and the 100-item cap, but omits the ecosystem default of 'npm' and gives no format hints for version or token, leaving some ambiguity the agent must resolve from 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?

The description implies a batch OSV lookup ('One OSV page per package') and the input shape, but never states the verb or resource explicitly — no 'query vulnerabilities for these packages'. An agent can infer the purpose but must read the schema to confirm it is a batch version of osv_query.

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?

No when-to-use guidance and no reference to sibling osv_query, which is the obvious single-package alternative. The only routing hint is 'at most 100', which implies batch use but never says to prefer osv_query for one package.

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, openWorldHint and destructiveHint=false, so safety is covered structurally. The description adds the genuinely useful fact that results are paginated and must be drained via a token, which is behavior beyond the annotations — but it calls the parameter 'next_page_token' while the schema names it 'page_token', which is a naming inconsistency an agent could misread.

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, zero filler, with the pagination constraint placed after the core action. Nothing extraneous and nothing buried.

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?

An output schema exists and the annotations cover the safety profile, so the description needn't explain return values. But with four parameters and zero schema documentation, leaving 'ecosystem' entirely unexplained and misnaming the page token leaves gaps an agent must guess through.

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, and it only gestures at name/version ('package version') and the page token — the latter under the wrong name. The 'ecosystem' parameter, which defaults to npm and materially changes results, is never mentioned in the description or 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?

The description states a specific verb and resource ('Fetch one OSV page') and scopes it to 'a package version', so the agent knows this is a single-page vulnerability lookup rather than a batch or CVE-database query. It stops short of naming which sibling to use instead (osv_vuln, osv_batch, nvd_cve), so the differentiation from the sibling list is only 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 second sentence gives a real usage rule — loop until has_more is false — which tells the agent this tool returns partial results. However, there is no guidance on when to pick this over osv_batch or osv_vuln, and no statement of what happens if the package is unknown or the ecosystem is omitted.

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 a useful note that package-specific affected ranges and fixed events are retained in the result, but says nothing about behavior on an unknown ID (empty vs. error) or about response size/latency for large records.

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?

One sentence, front-loaded with the verb and resource, with the qualifier about retained fields placed immediately after. Nothing is padded or 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?

An output schema exists, so the description needn't enumerate return fields, and the annotations cover safety. However, for a lookup keyed entirely on an ID string, the missing ID-format guidance and the not-found behavior are the exact gaps an agent would need filled before calling it confidently.

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 single parameter vuln_id carries no documented format. The description does not compensate: it never gives examples or states whether the ID is an OSV ID, CVE, or GHSA identifier. The name is largely self-explanatory, which keeps this from scoring lower.

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?

"Get an OSV record" names a specific verb and resource, and the qualifier about retaining package-specific affected ranges, fixed events, and references clarifies what kind of record is returned. It is clearly a single-record fetch, which implicitly contrasts with the sibling search tool osv_query, but the description never names that sibling to make the distinction explicit.

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 only implied: the presence of a required vuln_id signals "fetch one known record," which an agent can infer is the alternative to querying. There is no explicit when-to-use/when-not-to-use guidance and no mention of osv_query for lookups by package or version.

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