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

Annotations only declare readOnly, idempotent, and closed-world; the description adds substantive behavior on top: determinism ('the same question and dataset give the same packet_id'), provenance guarantees (every object carries a cite_as block with canonical URL and content hash), negative-space reporting ('what was left out and why'), and status semantics. These are exactly the traits an agent cannot derive from structured fields.

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?

Three sentences, front-loaded with the verb and scope, then determinism/provenance, then the status caveat. The enumeration of packet contents is dense but justified because it tells the agent what will come back. Slightly overstuffed for one paragraph but no filler sentences.

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 enumeration is technically redundant, yet the description still usefully covers determinism, provenance, and the INSUFFICIENT_EVIDENCE status an agent must branch on. Given a four-parameter read-only tool with full schema coverage, an agent has everything needed to call it correctly; only the sibling-routing context is absent.

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 all four parameters are already documented, and the description largely restates them ('as of a date', 'research question'). It adds the framing that the question is about the law rather than a person's case, but that constraint already appears in the schema's question description, so value beyond the schema is marginal — baseline 3.

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 ('collects') plus a precisely bounded resource ('evidence relevant to a research question about NYC sidewalk injury or traffic accident law as of a date'), and it enumerates the packet contents (propositions, statutes, cases, sources, relationships, conflicts, limitations, omissions). It does not, however, distinguish itself from siblings like search_evidence or get_evidence, so an agent still has to infer why it should build a packet rather than query evidence directly.

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 one concrete operating instruction — check the returned status first, and INSUFFICIENT_EVIDENCE signals the dataset does not cover the question — which implies when the tool is and isn't useful. It offers no explicit when-to-use-vs-alternatives routing against get_evidence, search_evidence, or get_relationships, so usage selection is only implied.

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