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

Annotations only flag destructiveHint=true and readOnlyHint=false, and the description goes well past that by disclosing versioning semantics (a new version is written, older versions are retained), scope ('writes only their own record'), recoverability ('they can undo it from this same surface'), and the confirmation model (runs immediately, returns an ordinary result rather than a plan_token). These are exactly the behaviors an agent needs and cannot get from the annotations. It stops short of stating permission requirements or limits on the new version history.

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 substance is front-loaded: what the tool does and the versioning rule come first, arguments second, confirmation last. The confirmation paragraph is the weakest part, since its closing sentence ('Everything destructive, outbound or financial still proposes and waits') describes the rule for other tools rather than this one, which is slight padding, but the overall length remains proportionate.

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 four-parameter mutation tool with no output schema, the definition covers the mutation's effect, version retention, undo path, and return shape ('an ordinary result, not a plan_token'). Nothing critical is missing for correct invocation, though it could say more about what the result contains and about permission prerequisites.

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 description coverage is 0%, so the description carries the burden and does name all four parameters, including that 'agreements' is a list of strings and which two are required. It does not explain what 'summary' semantically represents relative to clean_text, nor mention the size ceilings (maxLength 100000, maxItems 50), so the schema still contributes non-obvious detail. A solid compensation for the coverage gap, but not complete.

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 states a specific verb and resource: 'Correct a capture's recognised text', then immediately clarifies the mechanism ('Writes a NEW version on top (older versions are kept)'). This is far more informative than a restatement of the name, but it never distinguishes itself from near siblings such as website_update_text or intake_recognize, leaving the agent to infer why this intake variant is the right one.

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 tells the caller which arguments to pass and in what combination, and the confirmation paragraph explains that this particular write executes immediately rather than proposing. What is absent is any explicit when-to-use versus alternatives, e.g. that text should be corrected after intake_recognize has produced a capture, or when to prefer this over website_update_text. Usage context is implied rather than spelled out.

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