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

Annotations (readOnlyHint, destructiveHint=false, openWorldHint=false) already establish that this is a safe, local read operation, so the bar is low. The description goes further by disclosing method limitations (von Heijne and HMM-TM are weaker than SignalP/TMHMM), a concrete performance cost (DeepSig loads 9.8 MB and runs twenty networks per protein; deepsig is far slower than the matrix), and the deliberate omission of N-best decoding. It also documents the built-in self-check whose calls are false by construction, which is valuable behavioral context for interpreting results.

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 a wall of run-on sentences packed with parenthetical methodology. While the purpose is nominally front-loaded, the reader must parse a single sprawling sentence about conformal p-value formulas and predictor internals to extract operational meaning. The statistical detail crowds out the actionable guidance an agent needs first.

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 need not explain return values, and it instead supplies thorough methodological context: what a contradiction means, how p-values are computed and corrected, and the self-check. For a 7-parameter statistical tool with nested subject objects, this is close to complete; the only gap is that it never states prerequisites or the intended workflow position.

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 the baseline is 3. The description earns above-baseline credit by adding rationale the schema lacks: it explains why N-best decoding is excluded ('takes minutes per long protein, and the corpus has 5,231'), why BH is the default (conformal p-values against one calibration set are positively dependent), and the cost difference between the two signal predictors. It does not address _files or _subject, but those are covered by the schema.

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 opening clause gives a specific verb and resource: 'Find the signal peptides or transmembrane segments ... that a sequence predictor contradicts.' That is enough to separate it from sibling annotation tools such as protein_signal_peptide or protein_topology, which merely report predictions rather than testing UniProt against a predictor. The density of the rest of the sentence obscures the purpose somewhat, but the core action is unambiguous.

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?

There is no explicit 'use this when...' statement and no routing to or from the many related sibling tools. The closest thing to guidance is the interpretive caveat 'a disagreement between two models, to be read, not a correction of UniProt', which explains how to treat the output rather than when to invoke the tool. An agent must infer the use case from the methodology paragraph.

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