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

With annotations already declaring destructiveHint=true and idempotentHint=true, the description still adds substantial behavioral context: merge semantics for objects, spouse:{} vs spouse:null creating/removing, arrays replace wholesale (destroying unread records unless preserved), omitted fields unchanged, null/''/[] treated as unknown, and zero/false preserved. It also discloses auth requirements (authenticated One U Education account, declared planning scope, credentials never arguments) and the return contract. Minor gap: it doesn't explicitly describe expected_revision conflict/optimistic-concurrency failure beyond 'explicit validation errors'.

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 core action and the expected_revision contract are front-loaded, and nearly every sentence carries distinct operational information. It is dense to the point of being a wall of directives, with a few long multi-clause sentences that could be split for scanability, but there is little true filler.

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 destructive, open-world patch tool with an opaque payload and no schema descriptions, the definition covers what an agent needs: how to discover unknown shapes, how changes merge, how to clear values, what is preserved, auth/scope prerequisites, and that success returns the saved revision while failures return explicit validation errors. Output schema exists, so return-value detail beyond that is unnecessary.

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 0% and `changes` is an untyped open object, so the description does most of the work: it defines patch semantics for the changes payload (merge, null removal, array replacement, null-as-unknown) far beyond what the schema conveys. `expected_revision` is tied to the write but its conflict behavior is only alluded to, and `dossier_id` is never mentioned, so it falls short of full compensation.

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 opening sentence gives a specific verb (save) and resource (the user's explicit confirmed facts as structured changes at expected_revision), and the surrounding text implicitly separates it from the read-side siblings get_planning_dossier, get_planning_schema_index and get_planning_schema_definition. An agent can tell this is the write path for confirmed facts without opening the schema.

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 states explicit when-to-use conditions and exclusions: clarify ambiguous or inferred facts before writing, 'Schema discovery is not a prerequisite for every write', batch schema_refs into get_planning_schema_definition only when a shape is unknown, and 'fetch questions only when the next reply actually needs a new backend question, not as an automatic post-save check'. It also names the sibling tools to route to for schema lookups and states the confirmation gate before saving recommendations.

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