Detect vector outliers

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

The description discloses sampling behavior, clustering method (same as cluster_vectors), per-cluster scoring, sorting, and capping. It also notes the 'available=false' condition if pgvector is missing, which adds context beyond the readOnlyHint annotation. No contradictions.

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 description is a single, dense paragraph that efficiently covers functionality, algorithm, return values, and constraints. Every sentence adds value without redundancy.

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?

Given the tool's complexity (12 parameters, algorithm details) and presence of an output schema, the description thoroughly explains how it works, what it returns, and prerequisites. It addresses both parameter semantics and behavioral nuances.

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?

With only 8% schema coverage, the description provides extensive context for many parameters: sample_size, zscore_threshold, k, max_results, id_column, etc., explaining their roles and defaults. This significantly compensates for the sparse schema descriptions.

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 clearly states it flags rows with embeddings far from cluster centroids, explains the per-cluster z-score method, and distinguishes it from global outlier detection. It also mentions the return structure and includes the tool name's purpose.

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 description explains when to use (find outliers within clusters) and provides prerequisites (k >= 2, at least 2k rows, pgvector installed). It implicitly contrasts with a global approach but does not explicitly list alternatives or when not to use.

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