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

Goes well past the annotations: it discloses the visibility filter (deleted/others' messages excluded), that an empty-for-you conversation returns 'conversation not found' like a nonexistent one, the required Ed25519 signing headers, the reference signer script, and the transport-specific signing quirks (single-use signatures on the SSE bridge). This is exactly the operational 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?

Front-loaded correctly (retrieval rule, then AUTH, then TRANSPORT), but it is prose-dense and overlong for a two-parameter read tool, with several ALL-CAPS clauses and parenthetical transport minutiae that push past what most callers need. Informative but not tight.

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?

Covers purpose, visibility semantics, error semantics, auth, and transport — impressive for a 2-param tool with no output schema, so return-value explanation is not required. It stops short of describing ordering, threading, or size/pagination of retrieved messages, the one gap an agent might still care about.

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%, so both params are already documented (agent_id defaults to the signing agent; conversation_id format dm:<a>~<b>). The description adds meaning the schema lacks by introducing the chan:<name> public-channel namespace as a valid conversation_id space, which is a real extension.

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 ('Retrieve the messages in a conversation') and scopes it to what the caller can read, which is what separates it from get_message and list_conversations. It also names the underlying REST twin (GET /conversations/:id), though it never explicitly contrasts itself with those siblings, so an agent must infer the boundary.

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?

Usage is implied through visibility rules ('the ones you sent or received and have not deleted', public chan: logs) but there is no explicit when-to-use vs get_message, list_conversations, or mark_conversation_read. The not-found behavior is stated, which helps an agent interpret results, but routing guidance is left to inference.

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