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

Annotations already cover the safety profile (readOnly, idempotent, non-destructive), so the bar is lower, yet the description adds substantial context beyond them: sentiment is recorded only when enforce runs, zeros indicate off/shadow mode, low_sample rows should be flagged, and agents should be judged by recovery_rate rather than inherited negatives. This is exactly the extra behavioral context that annotations cannot carry.

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 purpose is front-loaded and most content (field semantics, enforcer/zero caveat, judge-by-recovery_rate rule) earns its place given the number of returned fields. However it is delivered as a single dense run-on with no structural breaks, and items like the Spanish naming rule ('sentimiento, never ánimo') and the list of frustration levels add bulk without clearly aiding tool invocation.

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?

There is no output schema, so the description must carry the return-value burden — and it does, enumerating summary, matrix, breakdown, frustration, flags, quality, and attention with their meanings and edge cases. For a complex no-output-schema report, the description is effectively complete for correct interpretation.

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% — days and the fields enum are fully documented in the schema itself. The description's detail ('conversations read, unread...') describes return fields rather than the input parameters, so it adds no meaningful semantics beyond the schema. 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?

The description states a specific verb+resource ('Customer sentiment over the last N days') and details the report's content precisely, so an agent knows exactly what it returns. It does not, however, name or differentiate itself from plausible siblings such as get_agent_report, get_conversation_report, or list_sentiment_conversations, so it stops short of a 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?

It gives downstream behavioral guidance ('Suggest reviewing the AI agent only when breakdown.agents has a bot row...') and an important interpretation caveat ('all zeros usually means it is off or still in shadow'), which imply when the report is meaningful. But it never explicitly states when to call this versus sibling reports, nor any prerequisite, so guidance remains implied rather than declared.

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