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

With no annotations, the description carries the full burden of behavioral disclosure, and it does so thoroughly. It explains the algorithm (greedy best-first, normalized similarity ratio), the threshold mechanism, return values (matched pairs with scores, unmatched indices), determinism, fixture verification, guest rate limits, the optional api_key benefit, and the Guild-signed provenance envelope. This far exceeds basic 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?

The description is structured from summary to algorithm to usage to schema, with each sentence contributing unique information. It is compact given the tool's complexity, avoiding filler while including necessary details like rate limits and provenance. The embedded schema is redundant with nothing else in the description and is essential for correct invocation.

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 tool with nested payload schema and no annotations, the description covers all critical aspects: what it does, how it works algorithmically, parameter details via schema, output behavior, determinism, authentication (api_key), and rate limits. The output schema is provided separately, and the description mentions the provenance envelope, ensuring the agent has enough context to use the tool 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 input schema only exposes api_key and a generic payload object, leaving the payload structure undefined. The description compensates by embedding the full JSON Schema for payload, specifying required fields, constraints (e.g., minItems, threshold range), and the meaning of threshold and key fields through prose like "caller-set threshold" and "chosen key fields." It also explains the api_key parameter's purpose. This fully makes up for the 0% schema_description_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 opens with a specific and actionable statement: "Fuzzy-match records across two lists by a key field." This clearly identifies the verb (fuzzy-match), the resource (records), and the scope (two lists), distinguishing it from siblings like ag_data_dedupe which likely operates on a single list. The later phrase "Entity-resolution lite" further positions the tool's purpose.

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 implies usage through phrases like "Entity-resolution lite: deterministic and auditable" and "free for guests," but it never explicitly states when to use this tool over alternatives or when not to use it. There is no mention of sibling tools like ag_data_dedupe for single-list deduplication, so the guidance is suggested rather than explicit.

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