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

Goes far beyond the read-only annotation by disclosing the internal refinement first search of then follow provision-to-case links then deduplicate by SAOS ID. Also reveals exact runtime behavior for rerank=false versus rerank=true, including satellite conditions such as Cohere reorderer config and the rerank_fetch_limit cap. This is strong behavioral transparency.

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?

Every sentence carries unique and decision-relevant information, and the opening line front-loads the primary purpose. The three-block structure moves cleanly from what to how to important cautions, with no purely ornamental text.

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 an output schema present, the description does not need to detail return values. It covers the discovery pipeline, rerank behavior, and behavioral limits. Slightly less complete for the full parameter matrix, but overall sufficient for an agent to invoke the tool with reasonable expectations.

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 parameter coverage is 0%, so the description must be nearly alone in explaining the 9 params. It meaningfully explains rerank, rerank_fetch_limit, and implicitly provision_limit via 'several provisions', but leaves candidate_limit, context_max_chars, case_text_max_chars, cases_per_provision, and rerank_result_limit undocumented. The star-heavy parameter semantics only partially compensates.

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 action and object: 'Discover related-case candidates for a natural-language legal issue'. It goes beyond the name by describing the discovery scope and clarifies that the tool does not decide holdings, which separates it from judgment-deciding tools like get_case.

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?

Provides clear situational guidance: this is a discovery stage, rerank=false is the inexpensive deterministic path, and successors should call get_case for any judgment used in an answer. It does not explicitly contrast itself with the sibling related_cases tool, so the guidance is not complete across all alternatives.

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