analyze_bill

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

With no annotations supplied, the description carries the full disclosure burden and does notable work: 'Read from local disk only; nothing is sent anywhere' establishes the privacy/safety profile, and conditional behavior is stated ('commitment coverage ... when the export carries commitment columns'). It does not address failure modes such as a non-FOCUS or malformed file, large-file behavior, or whether missing columns are skipped or error out.

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?

Front-loaded with purpose and privacy constraint, then outputs, then usage triggers — a sensible order for scanning. The output enumeration is dense but each clause names a distinct report section. Minor redundancy in asides like '(what discounts saved)' keeps it just under top marks.

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?

No output schema exists, yet the description compensates by enumerating the return payload (totals, grouped drivers, monthly breakdown, movers, list vs effective cost, commitment coverage). Since no annotations are present, the main residual gap is error/failure behavior and input validation expectations, which an agent would need before calling blind.

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 coverage is only 33%, so the description has to compensate. It partially does: the listed grouping dimensions map to the group_by enum, and the provider/file-format mentions supplement file_path. However top_n is never given meaning (it appears only indirectly via 'top cost drivers'), and neither its default of 10 nor how grouping and top_n interact is explained.

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?

Opens with a specific verb and resource: 'Analyse the user's actual cloud bill from a FOCUS 1.x CSV export', and enumerates supported sources (AWS Data Exports, Azure Cost Management, GCP BigQuery FOCUS view, OCI). It also lists the concrete outputs, so an agent knows precisely what it produces. It stops short of naming a sibling such as detect_bill_anomalies or analyze_cost_allocation to disambiguate, so it lands at 4 rather than 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?

The 'Use for:' block gives three concrete user questions this tool answers, which is strong positive routing. It also instructs 'Ask the user for the path to their export file', a prerequisite step agents often miss. There is no explicit when-not-to-use guidance or statement of which sibling handles adjacent questions like anomaly detection or re-pricing.

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