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

The description contributes substantial behavior beyond the annotations: sign-in and owner restriction, that the draft is rebuilt after the edit, that the corrected text replaces the whole source, and the publish-vs-private visibility model. However, it discloses that removing a source deletes an uploaded file — an irreversible destructive effect — while the annotations declare destructiveHint=false, so the two structured/unstructured signals conflict on the tool's most consequential behavior.

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?

Mostly dense and front-loaded — permissions first, then the two operation modes, then id sourcing and the data-loss caveat. The final sentence about shareable is ungrammatical and forces the reader to re-parse it, which is wasted space in an otherwise tight definition.

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, it covers auth, side effects (draft rebuild, file deletion), id sourcing, and visibility semantics, which is close to what an agent needs. The gap is that the ambiguous 'shareable' parameter and the destructive/annotation conflict leave one behavior under-specified in a way that matters for a write tool.

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%, so the description must carry all four parameters, and it largely does: source_id is where to get ids, text is 'pass the whole corrected text' (i.e. full replacement, not a diff), remove:true triggers deletion of an uploaded file, and shareable is tied to what the owner publishes versus what customers hear. Only 'shareable' is explained in broken phrasing ('shareable records that the owner is happy to share this source'), leaving whether it is a flag or a payload unclear.

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+resource pair and actually covers two distinct operations: correcting misread source text and removing the source. The 'e.g. a misread price' example and the 'one of the receptionist's sources' scoping make it easy to separate from add_receptionist_source or the generic 'receptionist' tool 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?

Gives explicit preconditions ('Owners only (needs sign-in)'), the situation that selects it ('Correct the text Latchbell read... e.g. a misread price'), and routes the agent to a sibling for prerequisites ('Get source ids from receptionist'). It does not say when to use add_receptionist_source instead for brand-new sources, but the guidance it does give is concrete.

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