send

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

Annotations say non-read-only, non-idempotent, non-destructive, but the description adds the substantive behavior: the recipient is woken, info=True suppresses the wake, files are copied into the room when sent, review answers are bounced back once if the peer omits file:line and severity, and due_min creates a 'late' report. This is exactly the context 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 return value are front-loaded in the first sentence, then per-argument notes. Mostly dense and useful, though a few parentheticals (the French 'relis'/'relecture' inference aliases, the chatgpt/agy glosses) are terser than they need to be for a reader with no project context.

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 seven-parameter tool with an output schema, the description covers every argument and the key side effects. The gap is that topic is optional with a null default and the description never says what happens when it is omitted, nor what errors or rejection look like when a peer fails the review round-trip.

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% and none of the seven parameters carry titles beyond bare names, so the description carries the full burden and does: it enumerates the valid 'to' values (claude, antigravity/agy, codex/chatgpt) that are absent from the schema, specifies the topic slug rules and the one-topic-per-project convention, and explains the semantics of info, due_min, files, and review.

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 ('Post a message for another agent in the TOPIC's room') plus the side effect ('the recipient is woken automatically') and the return value ('Returns its id (for wait_reply)'). This clearly distinguishes it from siblings like reply, inbox, and read.

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 selection guidance for flags: info=True means nobody is woken, review requests a file:line/severity answer, due_min triggers late tracking. It never explicitly contrasts send with the sibling reply tool, so an agent must infer when to answer an existing message versus start a new one.

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