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

The annotations already flag destructiveHint=true, openWorldHint=true, and idempotentHint=false, so the safety profile is partially covered. The description adds material context the annotations cannot: conversations are readable by every KEY in them and by the operator; a sealed pair is end-to-end except that the local bridge seals and opens it, and this connector cannot. That is genuinely useful disclosure for an agent deciding whether to send sensitive content.

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 action-by-action colon list is front-loaded and scannable, but the prose is dense and run-on, with clauses like 'a KEY that shares no SPACE or conversation with you gets it as a request' packed into an action description. It is effective but not economical.

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 an 11-parameter tool that multiplexes ten actions across a rich schema and an existing output schema, the description covers the key operational and privacy semantics an agent needs. Remaining gaps are parameter-level details that the schema already addresses, and return-value behavior is covered by the output schema.

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 55%, so roughly half of the eleven parameters carry inline descriptions (to, seq, body, about, sealed, reply_to). The description reinforces the semantics of to (one to fifteen peers), sealed (a sealed pair requiring a KEY that knows you), and body implicitly, but leaves peer_id, conversation_id, days, idempotency_key, action dependencies, and seq format mostly to the schema.

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 enumerates the ten actions the tool performs and describes each in operative terms (start, send, accept, decline, leave, clear, mark_read, block, unblock, set_retention), so the resource and the verbs are concrete. It does not name or differentiate itself from siblings, but the action list is dense and specific.

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?

Several actions carry their conditions: start takes one peer id for a pair or two to fifteen for a group; send requires an existing conversation, and replying to a request accepts it; declining tells nobody. This is clear guidance on when actions apply, though there is no explicit routing against sibling tools like schellingaf_messages or schellingaf_mailbox.

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