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

Annotations already say readOnlyHint/destructiveHint=false, and the description adds real behavioral context beyond them: invalid rows are marked, oversized plans are refused, replaces_receipt supersedes atomically, and source_touch_id must be preserved. It also discloses what is returned (proposal ids and a receipt) and that no extra approval rounds are needed, which is exactly the kind of context annotations can't convey. No contradiction with readOnlyHint since it explicitly says 'Saves nothing.'

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?

It front-loads the core purpose and includes a useful example, but the body is a dense run of jargon-heavy, semicolon-chained clauses ('Link new rows with unique create keys and exact {$ref:key} relationship references; no extra approval rounds for ids'). Informative but harder to parse than it needs to be.

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?

Given a 3-parameter tool with a nested proposal array, an output schema, and full annotations, the description covers purpose, constraints, and the review-then-commit lifecycle well. The only gap is that the dense delivery makes some of that coverage harder to extract, but nothing an agent needs to invoke it correctly 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, but the description adds genuine semantics: exact {$ref:key} relationship references for new rows, resolved UUIDs required for existing targets, and the atomic-supersession meaning of replaces_receipt. It reinforces rather than duplicates the schema, though a few sentences (e.g. 'no extra approval rounds for ids') are hard to map to a specific parameter.

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?

States a specific verb and resource ('Present ONE complete review of companies, contacts, opportunities...') and immediately clarifies 'Saves nothing,' which separates it from write tools. It implies the relationship to commit_reviewed_updates, though the phrasing is heavily jargon-laden, so an agent gets the gist without a crisp one-line summary.

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?

An explicit 'When to use' clause and an example steer the agent toward showing the request once before commit. It also names a concrete alternative behavior (chat corrections replace the review; don't repeat a visible app's save question), though it never states when NOT to call it beyond the implicit 'before commit' framing.

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