SKILL.md Lint

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

Annotations already mark the tool as read-only and idempotent. The description confirms this by stating it never raises protocol errors and is read-only/idempotent, adding value by detailing error return format. However, the description could elaborate on side effects (even if none) or behavior under unexpected inputs beyond malformed ones.

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?

The description is well-structured: it starts with a one-line summary, provides a concrete example, then usage notes. Every sentence adds value, and it's appropriately sized for the tool's simplicity.

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?

Given the tool's low complexity (one parameter, no enums, no nested objects, clear output schema), the description is complete. It covers input expectations, output format, error handling, and assumptions, with no gaps for the intended use case.

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%, so the schema already documents the parameter well. The description adds value by showing a usage example with expected output format, though it doesn't add much beyond clarifying input/output examples.

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 clearly states the tool explains a 5-field cron expression in plain English and flags mistakes. It differentiates from siblings (e.g., diff_texts, json_validate) by being narrowly scoped to cron scheduling, not general text or code validation.

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 explicitly says when to use it (to read a schedule) and what it's not for (6-field or seconds-precision variants). It also notes it's safe to retry after correcting input, but doesn't explicitly contrast with non-sibling alternatives or edge cases like timezones.

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?

The annotations already declare readOnlyHint=true, idempotentHint=true, and destructiveHint=false. The description goes well beyond by specifying that errors are never protocol errors but structured JSON, providing exact error examples, stating the diff cap (400 lines), describing the output format for identical inputs, and confirming read-only/idempotent nature. No contradictions.

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?

The description is well-structured: purpose first, then usage guidance, then error handling, then safety note. All sentences serve a purpose. Slight verbosity in the error example and the 'FREE.' label could be trimmed, but overall it is efficient and 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?

Given the tool's simplicity (2 params, no enums, output schema exists but not shown), the description fully covers input format, output format (both normal and error), limits (400 diff lines, 500000 char text), and behavioral properties (read-only, idempotent, retry safety). No significant 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?

Schema description coverage is 100%, with clear parameter descriptions already in the schema. The description adds a typical input example ({a, b}) and reaffirms the character limit in an error example, but these are supplementary rather than essential. Baseline of 3 is appropriate as schema does the heavy lifting.

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-resource pairing: 'Produce a unified diff between two texts.' It clearly distinguishes itself from a related tool (review_diff) that judges risk of code changes. Among the listed siblings, it is the only text comparison tool, making its purpose unmistakable.

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 explicitly states when to use ('show exactly what changed between two versions of a text'), when not to use ('not for judging the risk of a code change'), and provides an explicit alternative ('the codereview server's review_diff does that'). It also gives example input and explains retry safety, leaving no ambiguity.

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, and destructiveHint=false, so the safety profile is clear. The description adds value by detailing error behavior (returns error object, never a protocol error) and stating 'Every call is read-only and idempotent' — reinforcing annotations and adding concrete error-handling context. Minor gap: no mention of rate limits or auth (though not expected with FREE). Score 4 because annotations already cover most behavioral traits, but description enriches with practical details.

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?

The description is moderately concise and well-structured with front-loaded purpose and usage, followed by examples and error behavior. It is slightly verbose (e.g., redundant emphasis on idempotency), but every sentence provides useful information without significant waste. A minor improvement would be to remove the duplicate error behavior note around 'Errors: on invalid...'.

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?

Given the tool has a single parameter with 100% schema coverage, explicit annotations, and an output schema (which may exist but is not shown), the description is fully complete for the tool's complexity. It covers purpose, usage, error handling, idempotency, and sibling differentiation — no gaps detected for a read-only validation 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 100%, meaning the schema already documents the 'text' parameter well (type, minLength, description: 'The JSON document to validate, as a raw string; must be non-empty.'). The description adds an example of typical input ({"text": "{\"a\": [1, 2]}"}) and notes that the text must be non-empty, which aligns with schema. However, it doesn't provide additional parameter semantics beyond the schema — baseline 3 is appropriate given high schema coverage.

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 clearly states the tool validates JSON text and summarizes its structure, using specific verbs ('Validate', 'summarize') and resources ('JSON text'). It distinguishes itself from siblings like regex_test by explicitly noting it is 'Not for testing a pattern (regex_test)', and provides concrete examples of input and output.

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 explicitly states when to use the tool ('Use when JSON may be malformed or its shape is unknown'), what it reports ('structure, not conformance to a schema'), and a specific alternative ('Not for testing a pattern (regex_test)'). It also provides error-handling behavior and a safety note about idempotency, offering comprehensive 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 readOnlyHint, idempotentHint, and destructiveHint as true, false, false respectively. The description reinforces this by stating 'Every call is read-only and idempotent' and explicitly describes error-returning behavior instead of protocol errors. It adds detail about how errors are returned as structured JSON with fix instructions.

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?

The description is efficiently structured: a one-liner purpose, bulleted checklist of lint operations, typical input/output format, usage note, and error behavior. Every sentence serves a clear purpose without redundancy.

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?

Given the tool has an output schema (handling return values), the description covers all needed context: input format, use case boundaries, error handling, and safety guarantees. It is fully complete for a lint tool with good annotations.

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% with a clear description for the single 'content' parameter. The description adds value by explaining the input format ('full SKILL.md text including frontmatter') and typical retrieval from the tool's response, which slightly exceeds what the schema provides.

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 clearly states the tool lints one SKILL.md file, contrasting it with the sibling 'packaging_check' tool for archive layout. It specifies the exact checks performed (YAML frontmatter, required fields, description quality, etc.), making the purpose 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?

The description explicitly says 'Use on the full text of one SKILL.md. Not for the surrounding archive layout (packaging_check).' It provides typical input/output examples and error behavior, including that it is safe to retry after correcting input. This gives clear when-to-use and when-not-to-use 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?

Beyond annotations (readOnlyHint, idempotentHint, destructiveHint: false), the description discloses that the tool returns error messages instead of raising protocol errors, that it is safe to retry after correcting input, and provides full example error output. No contradictions with 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?

The description is efficiently structured: starts with single-line purpose and cost, then example, usage guidance, error behavior. Two minor areas reduce score: the example could be slightly leaner ('Typical input...') and the error notes repeat the JSON format. Still highly effective.

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?

Given 2 parameters, full schema coverage, and an output schema, the description largely completes the picture with usage guidance and error behavior. The output schema exists but is not referenced; describing return value types briefly could push completeness to 5.

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%, so the schema already documents both parameters thoroughly. The description adds a concrete input example but does not add additional semantics beyond the schema's property descriptions.

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 ('Validate'), resource ('zip layout plan for a skill marketplace'), and context ('before packaging'). It also distinguishes from the sibling tool lint_skill by noting it checks layout, not content.

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 explicitly says when to use ('on a planned file layout, before zipping') and when not to use ('Not for the content of a skill file – lint_skill'). This provides clear 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?

The description adds substantial behavioral context beyond the annotations: explains the CPU-capped child process to handle catastrophic backtracking, guarantees no protocol errors (returns error objects with fix guidance), and reiterates read-only/idempotent safety. This goes well beyond the readOnlyHint and idempotentHint 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?

The description is longer than typical but every section earns its place: purpose, example, usage, performance, error handling, and safety. It is front-loaded with the core function and example, and the ordering is logical. No verbose or redundant content.

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 tool with an output schema, the description covers all necessary aspects: what it does, how to use it, error handling, performance behavior, and retry safety. The output schema explains return values, so nothing else is missing. It is complete for an agent to call it 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?

The schema already provides 100% coverage for both parameters with meaningful descriptions. The description adds a typical input example that clarifies exact formatting (e.g., double backslashes in the pattern) and the result structure, which is valuable on top of the schema. The baseline of 3 is exceeded by this example.

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 ('Test'), resource ('regular expression against sample strings'), and provides a concrete input/output example. It also explicitly names json_validate as the tool for JSON validation, distinguishing itself from 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?

It gives clear when-to-use guidance ('Use to confirm a pattern matches what you expect before shipping it') and an explicit exclusion ('Not for validating JSON (json_validate)') with a named alternative. This fully routes the agent.

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