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

Annotations already establish readOnly/non-destructive/closed-world, so the bar is lower, and the description still contributes the write-permission gate and behavioral detail on the data itself (allowed-sites empty means any, 'expiresAt' null until removed, hidden items counted but not shown, 'script' emitted once per page). It omits ordering and pagination, but the authorization and data-visibility behavior are genuine additions beyond the annotations.

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 whole definition is one sprawling run-on sentence whose middle is a parenthetical field dictionary ('its id, "kind" ..., "name" ..., the ready code — ...'). The one actionable constraint, the write-permission requirement, is buried at the very end instead of being front-loaded, so structure works against 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?

With no output schema, the description does need to explain return shape and it attempts that comprehensively (kinds, ready codes, sites, expiry, creator, open counts, set items, hidden). But the field meanings are phrased so cryptically that an agent will struggle to map them to actual response keys, and ordering/pagination are unaddressed.

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% and the single parameter 'project' is documented in the schema with an example key ('e.g. MATH.'). The description says the scope is 'a project's ...' but adds no format, casing, or error behavior beyond what the schema already supplies, so the baseline of 3 applies.

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 does identify the resource — a project's embeds (terms/claims placed on other sites) and embed sets — which separates it from the create_*/update_*/delete_* embed siblings by listing intent. But the first clause elides its verb entirely ('A project's terms and claims put on other sites ... and its sets:'), so the agent must reconstruct that this is a read/list operation from the tool name rather than the prose.

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?

There is a real constraint stated — 'Only for somebody who may write in the project' — which tells the agent about authorization context. However, nothing says when to choose this over list_wiki_embeds, list_terms, or list_claims, and nothing warns about the empty-result case or what happens for a project the caller merely reads.

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