object_patterns

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

With readOnlyHint=true annotations, the description carries a moderate burden and delivers: it reveals the validate action is '<50 ms, offline', notes domain inference from other params, flags that domain is 'NOT a free-form pattern type', and describes the repair action as 'auto-fill missing required controls.' These specifics add real behavioral context beyond the annotations. The repair action is ambiguous about whether it mutates external state, but given the read-only toolkit framing this is not a clear 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 long but densely organized with a two-level hierarchy (domain → action) and consistent mode tags, front-loading the core 'choose a domain' concept. Given the tool's complexity — 12 parameters, 2 domains, 4 actions, nested recommend object — the length is justified and every section earns its place. Well-structured use of bold and bullets aids scanning.

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 high-complexity tool with no output schema, the description covers a great deal: domain inference rules, the recommendation decision flow, per-action return expectations ('Returns FP001-FP010 violations', 'full structure spec', 'a recommended pattern + reference forms to clone'), and when to call each action. Minor gaps exist around exact return formatting, but for the tool's breadth this is a thorough, near-complete description.

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 the baseline is 3 and the schema already documents every parameter with mode tags ([validate], [analyze], [spec]) and REQUIRED markers. The tool description adds workflow context that ties parameters together (e.g., recommend is 'preferred for a new form'), but the heavy lifting for parameter meaning lives in the schema's per-property descriptions, so it doesn't rise above baseline.

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 immediately establishes the resource (D365FO table and form patterns) with a specific verb ('Pattern toolkit. Choose a domain:'). It enumerates distinct modes (analyze, spec, validate) and even steers away from a sibling by stating that 'a concept like "number-sequence"/"SysOperation" belongs to get_knowledge, not here.' This clearly differentiates it from get_knowledge and other siblings.

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 explicit workflow sequencing: RECOMMEND for new forms, 'Call after analyze, before building' for spec, 'Call before action=create on d365fo_file' for validate. It also states a when-not rule by excluding pattern-type concepts that belong to get_knowledge. It doesn't exhaustively compare against every sibling alternative, but the internal action ordering and explicit get_knowledge exclusion provide strong usage guidance.

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