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

Far beyond the readOnlyHint annotation, it discloses the result cap (60), the priority ordering (Removed > Modified > Added), the filtering behavior (only objects touched by client code), the auto-fill behavior when params are omitted, the multi-model comma-separated handling, and the return shape (JSON payload with an 'instructions' field and 'objects' array). The agent knows what will come back and what to do with it.

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?

Every sentence earns its place and the description is properly front-loaded with WHEN, triggers, and sequencing before behavioral detail. It loses one point for being one dense unbroken paragraph — the content is warranted for this complex orchestration tool, but section breaks would improve scannability.

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 tool with 4 optional params, no output schema, and a multi-step workflow, the description is remarkably complete: it covers prerequisites (resolve_client_profile), fallback data source (list_release_note_inputs), guardrails ('never guess them'), output contract (instructions field + objects array), result limits and priority, and the exact follow-up call (generate_release_note_document). No critical information is missing.

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 coverage is 100%, so the baseline is 3. The description adds value beyond the schema by encoding the decision rule: omit v1/v2/customModelIds when a client profile exists, pass explicit values from list_release_note_inputs when it doesn't. It also reinforces the multi-model scenario (own extensions plus a separate ISV vendor model) and explains that businessContext 'sharpens the opportunity/regression assessment.'

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 names a specific verb+resource: 'Diffs two indexed D365FO versions... cross-references EVERY changed standard object against ALL the given custom models... returning ONLY the subset of changes actually touched by the client's code.' It also states its niche — 'you want to do the reasoning yourself instead of the server calling its own LLM' — which clearly separates it from broader siblings like analyze_upgrade_impact, diff_model_versions, and compare_objects.

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?

Explicit WHEN clause, concrete trigger phrases ('release note', 'regression risk', 'note de version'), and explicit sequencing: 'Call resolve_client_profile FIRST,' with conditional logic (omit params if a profile is found, otherwise 'call list_release_note_inputs to get real values and pass them explicitly -- never guess them'). It also names the downstream tool, generate_release_note_document, so the agent knows the full workflow.

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