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

With no annotations, the description carries the full burden and does deliver real behavioral facts: rate limit '≤600/hour/key → 429', latency target '≤400ms whisper', idle cost '$0', and the constraint that the tool performs no STT/LLM. It omits auth requirements, idempotency of repeat upserts, and what happens if the call is already ended, so it falls short of a 5.

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 core purpose is front-loaded in the first clause, which is good, but the body is a run of semicolon-joined fragments ('Path to ≤400ms whisper', 'Managed SaaS in-call index; no DID sales', 'Own PBX optional footnote') that are hard to parse. Little is wasted in length, but several fragments are marketing noise rather than instruction.

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 streaming, 9-parameter, live-call state mutation with no annotations and no output schema, the description covers context, rate limits, and cost but never describes the response shape or error semantics an agent would need after calling it. Since no output schema exists, adding return-value guidance would have been valuable and is absent.

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 signature line enumerates most parameters (call_id, text, speaker, t_ms, is_final, live, end_call, promote) but adds no meaning beyond the names, and pbx_id is not mentioned at all. With schema description coverage at 67%, the schema already documents several params, so baseline 3 is appropriate.

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+resource ('Upsert mid-call partial index') and what the input maps to ('utterances → fact/promise hints'), so an agent knows it accumulates hints from caller-posted text. It distinguishes itself reasonably from siblings like pbx_index_partial_get and pbx_ingest_transcript via 'customer posts text' and 'live call'. The telegraphic rate-limit and marketing fragments dilute an otherwise clear statement, keeping it below 5.

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?

'While the call is live' plus 'No Temsor STT/LLM — customer posts text' implies the intended operating context and contrasts with ingest-style tools. However, no alternative tool is named explicitly and no when-not-to-use condition is given, so the routing guidance is only implied.

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