billy_import_receipt

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

Annotations already declare readOnlyHint=false, destructiveHint=true, and idempotentHint=false, so the agent knows it is a mutating, non-idempotent operation. The description adds useful behavioral context: 'Same bytes are deduplicated' explains a key idempotency-like behavior, and 'Does not upload to Billy' clarifies a side effect that is not in the annotations. This goes beyond the annotations without contradicting them.

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 three sentences with zero waste. It front-loads the primary action and then adds crucial qualifiers (dedup, non-upload) in a logical order. Every sentence adds value, and the structure is easy to scan. This is exemplary conciseness.

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 (3 required params, nested source and metadata objects, many required metadata fields, no output schema, and annotations), the description covers the main purpose and a couple of behaviors, but it leaves gaps. It does not explain what the return value is (if any), what 'configured inbox' means, or what happens if a file cannot be found. It also does not elaborate on the expected format of metadata or the significance of source.kind beyond the enum. The presence of a rich schema partially compensates, but the description itself is not fully complete for a tool with this many 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?

Schema description coverage is 0%, so the description must compensate. It does mention 'extracted invoice metadata and provenance' which hints at the 'metadata' and 'source' parameters, and the enum of sources (Gmail, Drive, local, vendor portal) is reflected. However, it does not explain the required fields inside metadata (supplier, invoiceNumber, amounts, etc.) or the meaning of filePath and source fields. The high-level hint is useful but insufficient for a parameter-heavy tool with nested objects.

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 action ('Archive an original PDF/PNG/JPEG'), a clear resource (the original document), and a scope (from the configured inbox). It also mentions the key outcome (extract metadata and provenance) and distinguishes itself by noting 'Does not upload to Billy', which sets it apart from potential upload-related siblings. The purpose is unambiguous and well-scoped.

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 implies usage: it is for archiving a document with extracted metadata and provenance from listed sources. However, it does not explicitly state when to use this tool versus alternatives like billy_receipts (likely a list operation) or billy_prepare/execute (likely processing steps). The only hint is the negation about uploading, but no direct comparison or exclusion is given. This is clear context but lacks explicit routing guidance.

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