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

Annotations only declare the write/safety profile; the description goes well beyond by disclosing the overlay model (source rows never overwritten, edits merged only under overlay="with_overlay" with original value and provenance), the pending-proposal lifecycle (counts nowhere until the owner accepts or rejects), the return value (proposal ids), and the API key requirement.

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 scope are front-loaded in the first two sentences, and the workflow follows logically. There is some redundancy: the overlay rule is stated twice (source rows never changed, then the explicit 'Overlay rule' sentence) and the row_key instruction mirrors the schema description.

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 still states what is returned (proposal ids). Combined with auth requirements, supported connector tables, the row_key prerequisite and the approval lifecycle, an agent has everything needed to invoke it correctly.

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%, including the row_key sourcing note, the column naming rules and the per-entry vs. top-level `reason` fallback, so the schema already carries parameter meaning. The description largely restates the row_key sourcing rather than adding new semantics, making 3 the correct baseline.

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?

Names a specific verb (propose) plus the exact resource (a DERIVED column on a dataset) and illustrates what such a column holds: topic, group, label, campaign, cleaned category. It is clearly separable from siblings like propose_corrections, propose_flags and propose_mappings, which do not create a new column.

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 usage context ('Use it to label or group posts, orders or campaigns') and the operational prerequisite of sourcing `_row_key` from query_dataset with overlay="with_overlay". It also points to the accept path (accept_proposals / dataset page) for the downstream step, though it never explicitly contrasts itself with the sibling propose_* tools.

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