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

With readOnlyHint and openWorldHint=false already declaring a safe, local operation, the description still adds meaningful context: 'run on this machine' confirms local execution, and 'one PSI-BLAST iteration' plus the SMART-database profile construction disclose the actual computational behaviour. It stops short of warning what happens when the protein has no SMART domain hits (the obvious failure mode) or of any runtime expectation.

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?

A single tightly packed sentence, front-loaded with the tool's name and the algorithm sequence, with the em-dash used to separate the pipeline (profile then search) from the rationale. Nothing is repeated from the schema or annotations, though the sentence is dense enough to require re-reading.

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 9-parameter tool with annotations and an output schema, most of the burden is already covered by structured fields, and the description correctly explains the method rather than the return values. What is missing is the failure/edge case: a protein with no SMART conserved domains cannot build a profile, and an agent has no guidance on that outcome. Runtime and the cost of the single iteration are also unstated.

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 coverage is 100%, so each of the 9 parameters (evalue, threshold, domain_inclusion_evalue, max_target_seqs, source, protein, subjects, _subject, _files) is already documented in the schema. The description adds conceptual meaning by tying the run to profile construction from conserved domains, but it adds no per-parameter guidance beyond that. Baseline 3 applies.

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 method (DELTA-BLAST), a specific resource (a protein's SMART conserved domains) and the exact algorithm (build a PSSM, then one PSI-BLAST iteration), plus a sensitivity claim versus BLASTP. An agent can distinguish it from a plain search or a domain-scan. It does not name its closest siblings (e.g. blast_search or the rps_blast variants), so differentiation is by method description rather than explicit routing.

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 closing clause 'more sensitive than BLASTP for distant homologs' is an implicit when-to-use signal, and the schema notes that an empty subjects value searches the open record's CDS. There is no explicit statement of when NOT to use it (e.g. for close homologs or nucleotide queries) and no named alternative tool to fall back on.

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