Traiter un justificatif en note de frais

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

Despite the readOnlyHint annotation, the description adds important behavioral context: it handles illegible fields by asking the user, it never invents receipt amounts, and it returns a text runbook that the model must later execute. The only slight ambiguity is the phrase 'crée la ligne de frais' against the readOnlyHint, but the following sentence clarifies that the tool actually returns a runbook to be executed. No hard contradiction.

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 compact, French-first, and front-loads the core purpose. It includes necessary workflow details, user validation, and actual output without unnecessary fluff. The short note about MCP client compatibility earns its place by explaining why this tool exists.

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?

With no output schema, the description clearly states the output is a runbook to execute by calling Boond tools. It covers input, validation rules, no-invented-amount behavior, and the downstream action. It does not detail the runbook format, but for a workflow tool this is sufficient.

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?

All four parameters are fully described in the input schema (100% schema description coverage), including defaults, resolution logic, and use of alternative search tools. The main description itself does not add significant extra parameter-level semantics, so a baseline of 3 is appropriate.

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 uses specific verbs: extracts expense data and creates the corresponding expense line (via a follow-up runbook). It clearly states the input (receipt photo/PDF), the scope (BoondManager expense report), and uniquely identifies the workflow, distinguishing it from other sibling workflow tools.

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 explains when to use it: when a photo or PDF of a receipt is attached. It also gives a strong behavioral guideline: validate explicitly with the user before proceeding, and it positions itself as a tool equivalent to the MCP prompt for clients that don’t handle MCP prompts well. It does not explicitly list exclusions or direct alternatives, but the conditions are clear.

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