Check what is waiting for you

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

Annotations only declare readOnlyHint false and idempotentHint true; the description discloses the actual mutation mechanism ('Peeks unless drain=true'), the exception ('needs='human' items are never drained, since only a relayed human answer clears one'), cost behavior ('Costs a fetch per unfollowed room'), conditional defaults ('Defaults true only while you have nothing in flight'), ordering ('in flight first, then stalled'), and non-switching reads across networks ('Reading one never switches it'). This is substantial behavior beyond what annotations provide, and it is consistent with them — no contradiction.

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?

There is zero filler and the default scope is front-loaded, but the prose is telegraphic and run-on — 'Peeks unless drain=true; needs='human' items are never drained, since only a relayed human answer clears one' packs multiple behaviors into one compressed sentence. The three scopes run together in a stream, reducing parseability for an agent.

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?

Behavioral coverage is strong and scope semantics are well defined, but the tool has no output schema and the description never states the return shape — what fields or format the peek returns. Additionally, two of seven parameters (room, limit) remain undefined. For a 7-parameter tool with no output schema, these are material gaps.

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 descriptions cover all 7 parameters but are cryptic one-word pointers ('pending', 'owed', 'Default 'pending''). The main description adds real meaning by defining the three scope values the schema references and by elaborating drain, needs, includeUnclaimed, and network. However, room (schema description: 'pending') and limit (schema description: 'owed') are never explained in either place, so their semantics must be inferred.

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 title 'Check what is waiting for you' supplies the verb, and the three scope definitions — 'pending (default): messages addressed to you, not yet processed', 'owed: every unfinished task you are assigned, were offered, created, or could claim', 'unrouted: messages naming you in rooms you never joined' — make the inbox-listing role discernible and distinct from siblings like komnet_read or komnet_wait. However, the purpose is never stated directly as a sentence (e.g., 'returns the list of items waiting for you'); it is conveyed entirely through scope definitions, with 'Peeks' as the only explicit verb.

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?

One explicit use case is given for the unrouted scope ('so use it when someone says they sent you something you never saw') plus a cost warning ('Costs a fetch per unfollowed room'). But no alternative tools are named, no when-not-to-use is stated, and usage for the default 'pending' and 'owed' scopes is implied by their definitions rather than spelled out.

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