Dependency check

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 profile is covered. The description adds genuinely useful behavior: the exact returned fields, the free-call quota (5 per UTC day per IP), the X-API-Key requirement after that, per-call credit cost, and that not_found and fetch_failed are neither charged nor consume a free call. That is above the annotation bar, but it omits response shape details and whether results are cached or versioned.

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-loads purpose and return fields before billing mechanics, which is the right order. It repeats the full ecosystem list already present in the enum, and the signup URL sentence is promotional rather than functional, but overall it is 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?

With no output schema, the description carries the return-value explanation and does so precisely (advisory id, summary, CVE aliases, severity, fixed versions). Error-state and billing semantics are also covered, leaving nothing an agent needs in order to call this 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 coverage is 100%, and the description largely restates what the schema already documents, including the ecosystem enum and the vendor/package default heuristic. It adds no format or syntax guidance beyond the schema, so the baseline of 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: "Look up known vulnerabilities for one package version from OSV." It also scopes itself with "This is not a full audit." It does not name the sibling check_package or explain how this differs from it, so an agent must infer the boundary.

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 operating context: single package version, empty list means OSV returned nothing, and successful empty lookups are still charged. The lack of explicit comparison against check_package keeps it from 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 already cover safety (readOnly, idempotent, non-destructive, open world), and the description adds substantial context they cannot carry: the free-tier quota (5 successful calls per UTC day per network address), which outcomes are uncharged (not_found, fetch_failed), the requirement to send X-API-Key afterwards, and the 0.04 credit cost on success. It also discloses the conditional return behavior for a missing version, which is exactly the kind of behavior an agent needs before calling.

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 purpose and outputs, then scoping, then the commercial mechanics; almost every sentence carries operational weight. The signup URL and credit pricing are necessary but make the tail denser than ideal.

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, so the description has to describe return values and it does: the four facts plus the latest-stable fallback and version_found. It also covers failure modes (not_found, fetch_failed) and their billing treatment, leaving nothing an agent needs for a correct call unstated.

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 baseline is 3, but the description adds real meaning: 'Send vendor/package' clarifies the expected identifier shape, and it explains the version parameter's effect plus the version_found=false outcome when a release is absent while the latest stable is still returned. That goes beyond the schema's terse 'Optional release to select.'

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 (current Packagist facts for one PHP package) and enumerates exactly what comes back: latest stable version, required PHP, license, abandoned status. It does not explicitly contrast itself with the sibling check_dependency, 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?

Gives a concrete trigger point ('Use it before composer require') and an explicit scope exclusion ('This does not list vulnerabilities'), which tells the agent when not to reach for it. It never names check_dependency as the alternative for dependency-level questions, so the routing decision between siblings is left to inference.

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