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

The description discloses substantial non-annotation behavior: it applies sign-appropriate detail fields, writes user overrides that AI enrichment will not overwrite, re-derives journal entries, and returns per-transaction failures. This goes well beyond the idempotentHint and destructiveHint annotations, giving the agent a real picture of the side effects and the persistence guarantees of the operation.

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 the core operation front-loaded, followed by side effects and then a usage alternative. Every sentence contributes a distinct, valuable piece of information with no filler or redundant praise.

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 batch mutation tool that writes overrides and re-derives accounting entries, the description covers the important aspects: scope, side effects, return information, and a sibling alternative. It also includes the key constraint of 500 transactions. It is missing only explicit notes on whether failures are partial or atomic and how this tool relates to the by-account batch variant, but given the output schema and annotations, this is sufficient for an agent to invoke it.

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?

The schema description coverage is only 33% (only company wrapper has a minimal description) but the description partially compensates by explaining the categoryOptionCode with concrete examples and clarifying the batch limit for transactionIds. It does not deeply explain the companyId or the wrapper objects, leaving some parameter semantics to naming conventions and schema structure.

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 the core action clearly: set the same category on up to 500 transactions in one call, with the categorization key type (taxonomy option code) and examples. It distinguishes itself from the single-transaction sibling by explaining that this is the batch alternative, making the tool's intent unambiguous even though the by-account sibling is not explicitly mentioned.

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 gives an explicit usage directive: 'Use this instead of calling transaction_categorize_transaction in a loop.' It clearly implies the primary use case (same category across many transactions) but does not explicitly cover exclusions such as when different categories are needed per transaction or when to prefer the by-account batch sibling. The guidance is helpful but not exhaustive.

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