beel_list_invoices

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

Annotations already include readOnlyHint=true, idempotentHint=true, destructiveHint=false. The description adds the key behavioral fact that only documents of the company in the path are returned, and scopes the account context (company_id is the only source of context; header plays no part). It also warns about fiscal guardrails (when an invoice can still be changed) and directs to docs for exhaustive rules. This adds value beyond the annotations, though it doesn't mention pagination details or rate limits explicitly, but the baseline is lowered by strong annotations. No contradiction found.

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 short, front-loaded with the core purpose, then the endpoint, then a pointed warning about fiscal guardrails and a pointer to docs. Every sentence earns its place; the guardrail note is specifically actionable. No redundancy with annotations. Structure is clear and concise.

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 complex 25-parameter list tool with no output schema, the description gives the essential context: scope, key filters, and a critical safety pointer (guardrails). It does not enumerate every filter but the schema already does that (100% coverage). The description handles the company scoping nuance and directs to docs for exhaustive rules, which is appropriate given the tool complexity. Missing pagination specifics (like default limit, max) are minor and covered by schema defaults. It is sufficiently complete for an agent to call correctly.

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 100%, so according to the rubric, the baseline is 3. However, the description adds value by summarizing the key filter categories (status, type, series, customer, date range, free text) which helps the agent see the breadth without reading every parameter. It also mentions 'free text' which maps to the `search` parameter and highlights the `fiscal_only` concept implicitly via 'invoices of this company'. The description does not repeat parameter details—that's left to the schema—but it provides a useful high-level summary. Since coverage is high, and the description adds a compact view of the filters, a 4 is justified.

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 ('Returns'), resource ('paginated list of invoices'), scope ('this company (NIF)'), and lists filterable dimensions (status, type, series, customer, date range, free text). It also explicitly scopes results to only the company in the path, clearly distinguishing it from any global list. The endpoint is named, and the guardrail reference adds context. This clearly separates it from siblings like beel_list_invitations and other list 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?

The description clearly states when to use: to list invoices of a specific company, with extensive filtering options. It does not explicitly mention when not to use alternatives, but the scoping to a single company and the guardrail note about invoice state machine provide context. The mention of 'fiscal guardrails' and reference to beel_docs_search for exhaustive rules gives usage direction, though there are no explicit exclusion criteria against other list tools. This is better than many because it points to a docs search for edge cases.

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