aurora_analyze

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

With no annotations provided, the description carries the full burden of behavioral disclosure, and it does so richly: read-only, local, on-device execution, cited findings with claim_id, contractually zero fabricated_count, changepoint calibration blocks, and verdict downgrade behavior. It does not discuss error conditions or limits, but it goes far beyond a minimal description. No contradiction with annotations because none exist.

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 dense but well-structured: purpose and formats lead, followed by methods, return characteristics, and finally usage guidance. Each sentence carries substantive information; the only mild redundancy is the closing list ('analyze data, find anomalies, check what changed'), which is acceptable as a routing hint. It is longer than typical, but length is justified by the richness of the tool's behavior.

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 no output schema and no annotations, this description covers input types, return payload (cited findings, confidence, fabricated_count), behavioral guarantees (calibration, downgrade), and usage routing. The only notable omission is parameter semantics for depth and sections. For a statistically complex tool, this is strong coverage — an agent can invoke it correctly without opening the schema.

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 only 25%, meaning depth, sections, and full_bundle are undocumented in the schema. The description clarifies full_bundle ('compact summary unless full_bundle=true') and partially clarifies path ('CSV path, existing run_dir, or .aurora.json'), but it does not explain the semantic meaning of depth (auto/quick/standard/full) or sections. The description compensates somewhat but not enough to fully bridge the parameter meaning gap.

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 ('Statistical analysis'), a clear resource ('dataset CSV, TSV, Parquet, XLSX'), and enumerates concrete capabilities (anomaly detection, changepoint detection, forecasting, causal discovery). It strongly distinguishes this tool from siblings by positioning it as the primary analysis entry point, name-dropping methods that no other sibling claims.

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 provides explicit when-to-use guidance: 'Use this FIRST whenever a user asks to analyze data, find anomalies, check what changed, or wants real statistics instead of estimates.' This directly tells the agent when to invoke this tool and implicitly routes around alternatives like aurora_forecast or aurora_simulate. This is exactly the kind of usage context an agent needs.

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