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

With no annotations, the description carries the full burden and does so well: it discloses nucleus-dependent coverage (1H full, 13C only 30 entries in CDCl3/DMSO-d6, standards carry no 13C), the error path (`unsupported_nucleus` on a nucleus with no data), and the crucial score-interpretation behaviour (self-match >= 0.800, random carbon lists average 0.24 and reach 0.625). This is exactly the operational context structured fields cannot convey.

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?

Front-loaded with purpose and scope, then coverage, then scoring semantics — a sound ordering with no filler. It runs long across three blocks, and the Fulmer citation plus exact self-match numbers are dense, but each block carries decision-relevant information rather than restating the name.

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 values (including `nucleus_coverage`) need not be described, and the interpretive guidance is unusually complete. However, for a tool whose only required argument is an input file path, failing to state the accepted spectrum format or the auth requirement implied by `api_key` leaves a real gap before an agent can call it successfully.

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 0% across 6 parameters, so the description must compensate and largely does not. It clarifies nucleus coverage and solvent constraints indirectly, but never explains the required `path` parameter's accepted format (peak list, JCAMP, raw string?), nor what `tolerance_ppm`, `max_candidates`, or `api_key` mean or how defaults behave. The single most important parameter, the input spectrum, is left undefined.

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 verb and resource (identify a spectrum against a curated reference set) and immediately bounds the scope: 39 entries, 7 qNMR standards, 32 impurities/solvents. It explicitly distinguishes itself from a general or natural-product database, so an agent knows exactly what class of problem this solves.

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?

Gives explicit when-to-use (confirming a known standard or common impurity), explicit when-not (unknown natural products, novel structures), and names both alternatives with the condition that selects each: covnmr_verify to test a suspected structure, covnmr_rank_candidates to choose between suspects.

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