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

Annotations declare readOnlyHint=false, destructiveHint=false and openWorldHint=false, and the description adds genuinely new behavior: creation is a proposal governed by the project publication policy, it returns change_id and change_status, and idempotency_key makes exact retries safe with a replay returning the original response plus replayed=true. The idempotentHint=false annotation is not contradicted, since the tool is only made safe by supplying a key rather than being inherently idempotent — the description explains precisely that nuance. It stops short of describing what happens to the created document pending approval or what errors can occur.

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?

Three sentences, all front-loaded: required vs optional fields first, then the publication-policy behavior, then return values and retry semantics. Almost every clause carries information, though the paraphrase "size limit" and "initial content" add a translation step when the schema uses max_chars and content.

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 15-parameter mutation with no output schema, the description usefully discloses the return shape (change_id, change_status) and the replay contract, which an agent would otherwise not know. It is nonetheless incomplete: a required parameter (project_id) is never mentioned, and five optional parameters (evidence, excludes, rationale, source_name, result_offset) have no meaning attached anywhere. Adequate to call the tool correctly for the common case, but not complete.

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?

With schema description coverage at 0% across 15 parameters, the description must carry the parameter burden, and it only partly does: it names title, purpose, update_when, instructions, format_template, load, max_chars ("size limit"), content ("initial content") and idempotency_key. It omits the other required parameter project_id entirely, and leaves evidence, excludes, rationale, source_name and result_offset unexplained. The coverage it does provide is helpful but the gap is material for a 15-parameter tool.

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 opens with a specific verb and resource: "Add a new document to a project," which clearly separates it from read_document, archive_document and restore_document among the siblings. It also names the substantive fields being set, so an agent knows this is a document-authoring call rather than a change/version operation. It does not explicitly name any sibling as an alternative, which keeps it short of a 5.

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?

Usage is implied rather than stated: the phrasing "proposed under the project publication policy" signals this is the entry point for new document creation and that the result goes through an approval flow. However, the description never says when to use this instead of amend_changes, submit_changes, or restore_document, nor does it state prerequisites such as project access. An agent can infer the right context but is given no explicit routing rule.

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