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

Annotations only declare non-readonly, open-world, non-idempotent, non-destructive; the description adds the genuinely new facts: nothing is published, a draft id exists for a later create_recipe call, and two named result states ('existing', 'blocked') with the required agent response for each. That is exactly the kind of workflow and side-effect disclosure annotations cannot carry.

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?

Purpose and the no-publish guarantee are front-loaded in the first two sentences, and the trailing sentences are all load-bearing state-handling rules. The prose is dense and clause-heavy ('existing means ... blocked means ...') which slows parsing slightly, but nothing is redundant.

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 4-parameter, open-world draft-creation tool with an output schema, the description covers the full lifecycle an agent needs: inputs (text/url/photo read-through), what the draft is, the confirmation-and-publish next step, and behavior for the two non-happy-path outcomes. No gaps remain that would cause a miselection or misequence.

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 conditional usage the schema only hints at: add_anyway is gated on the cook asserting it's a different recipe, and source_url is prescribed specifically for the paywalled/'blocked' case. It does not add format detail for url vs text beyond the schema's own 'pass text or url, not both'.

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 ('turn a recipe ... into a draft') and enumerates what the draft contains: matched ingredients, plain-word steps, servings, times, timers. It explicitly separates itself from the publish path ('Nothing is published ... publish with create_recipe({ draft_id })'), so an agent can distinguish it from create_recipe and save_recipe without opening either 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?

Gives concrete trigger utterances ('save my grandma's meatballs', 'add this one' with a link, a photographed card) plus the handling rule for the photo case (read it yourself and pass the text). It also routes on outcomes: 'existing' → send them to save_recipe/fork_recipe, add_anyway only if it's a different recipe, 'blocked' → pass page text with source_url.

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