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

With annotations already covering read-only and idempotent semantics, the description adds real behavioral context: a confidence score is returned, requests are backed by Jev with an automatic Groq LLM fallback, and latency is bounded at under 4.5s. That fallback and latency guarantee is exactly the kind of operational detail 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?

Purpose is front-loaded in the first clause, followed by backend/latency detail and a sample-endpoint pointer. Two compact sentences, no filler. The sample route is slightly incidental but cheap and useful.

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 documentation is not the description's burden. For a single-parameter text classifier, the description covers output classes, confidence, backend behavior, and latency, which is sufficient for correct invocation. Only cross-tool disambiguation is missing.

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% and the only parameter ('text') is documented in the schema with its 4000-character limit. The description adds no further parameter semantics, so the 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 verb ('classify') plus the resource and the exact output taxonomy: toxic vs not toxic, with hate speech, harassment, threats, and severe offensive language as the covered categories. An agent immediately knows what the tool returns. It does not, however, distinguish itself from close siblings like spam_check, sentiment, guard, or classify.

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?

Usage is implied by the purpose ('classify text as toxic') and by the sample-endpoint hint, but there is no explicit when-to-use guidance, no exclusions, and no routing away from the overlapping categorization tools (spam_check, sentiment, guard, classify). Adequate but leaves selection to inference.

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