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

No annotations are provided, so the description carries the full behavioral burden — and it does: ANN approximation with measured recall (~0.96), eventual consistency with a concrete 5-minute replica-refresh window, the precise nature of the variance (visibility of new observations, not ranking of visible ones), and a 30-day default time bound echoed in metadata.time_range.

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?

Well front-loaded — purpose in the first sentence, caveats next, usage and returns after — and each section is doing work. It is longer than strictly necessary, and the two worked examples largely restate the WHEN TO USE bullets, but the structure makes it scannable rather than sprawling.

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?

Although no output schema exists, the RETURNS section enumerates the response fields (observation_id, similarity, confidence, evidence_grade, metadata, suggested_next_queries), so an agent knows what to expect. Combined with the consistency caveats and time-boundary, nothing needed to call this correctly 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%, so the baseline is 3. The description adds value beyond the schema by stating the time-bound behavior ('searches the last 30 days by default. Pass filters.time_range to widen or narrow it') and the default limit of 20, plus noting that the applied window is surfaced in metadata.time_range.

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 with mechanism: 'Search observations by semantic similarity' using vector embeddings on observation payloads. The scope (promoted observations, ANN cosine over a Lance IVF_PQ index) is precise. It does not explicitly distinguish itself from close siblings like query_observations, find_similar_moments, or semantic_audience_search, so an agent must infer the boundary.

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?

Contains an explicit WHEN TO USE block with four concrete scenario types, plus a hard exclusion: 'Do not use this tool where a repeatable, exhaustive result set is required.' That is when-to-use, when-not-to-use, and the tradeoff that drives the choice.

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