detect_anomalies

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

The Returns section discloses the output shape (JSON with anomaly list and summary). It documents the deterministic nature of the tool ('deterministyczny, powtarzalny wynik'), which is a key behavioral trait. With no annotations provided, the description carries the burden and reasonably discloses the detection methods and return format, though it doesn't specify error conditions or data format constraints.

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 well-structured with Args and Returns sections, front-loading the purpose and rationale before parameter details. It's compact with no wasted sentences, though the rationale sentence about LLM weaknesses, while useful, could be considered slightly extraneous to purpose.

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 tool with 6 parameters, low schema coverage, and no annotations, the description provides strong coverage: purpose, both methods, defaults, return format, and the statistical rationale. The output schema exists and is referenced. Minor gaps: exact accepted data formats (CSV/JSON shape) are only mentioned briefly, and error handling isn't discussed.

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 0%, so the description must compensate. It explains each parameter: data, method (with both options and what each computes), sensitivity (threshold in std deviations, default 3.0), window (with auto behavior at 0), and ts_field/value_field ('nazwy kolumn'). This adds substantial meaning beyond bare schema titles, though exact formats for data strings are only loosely implied.

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 the tool detects anomalies in time series (outlier points), uses a specific verb+resource structure, and distinguishes it from siblings (e.g., detect_threshold_breaches for a different anomaly type). The rationale for use (LLMs poorly compute statistical deviations on thousands of points) adds real 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?

It explicitly explains WHEN to use this tool (when deterministic, repeatable anomaly detection is needed) and contrasts it with the LLM's weakness. It doesn't explicitly name alternatives like detect_threshold_breaches to disambiguate, but sibling tools are clearly different enough. The window auto-sizing note provides practical guidance.

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