Contract Clause Extractor

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

With no annotations provided, the description must carry the full burden of behavioral disclosure. It successfully adds context about the severity ratings system for risk flagging, which is not evident in the schema. However, it fails to disclose whether the tool is read-only (safe to call), what output format to expect (structured vs. text), or any rate limiting concerns.

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 with the primary action front-loaded, followed by enumerated examples and optional functionality. It avoids excessive verbosity. Minor deduction for 'Powered by Claude' which provides implementation trivia rather than selection-relevant information, and for 'and more' which is slightly vague (though mitigated by the schema enum).

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 absence of an output schema and annotations, the description adequately covers input parameters but leaves a significant gap regarding the return value structure—critical for an extraction tool. It does not indicate whether results are returned as JSON, markdown, or plain text, nor does it describe the structure of the severity ratings mentioned.

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?

Despite 100% schema coverage (baseline 3), the description adds meaningful semantic value by listing representative clause types that map to the enum values in the 'clauses' parameter, and crucially elaborates that 'flagRisks' produces 'severity ratings'—a behavioral detail absent from the schema's generic 'Flag risky or unfavorable clauses' 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 clearly states the specific action (extract) and resource (contract clauses), enumerating concrete examples like parties, payment terms, and liability. However, it lacks explicit differentiation from the sibling tool 'extract_from_text', leaving implicit the distinction that this tool is specialized for legal contracts versus general text extraction.

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?

While the domain (contracts) is specific enough to imply usage, the description provides no explicit guidance on when to use this tool versus alternatives like 'extract_from_text' or 'compare_documents'. There are no stated prerequisites, exclusions, or conditions that would help an agent decide between this and similar extraction capabilities.

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