inbox

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

This is the standout dimension. Annotations only say readOnlyHint=false, destructiveHint=false, openWorldHint=false, leaving the reason for the non-read-only hint opaque. The description explains it: a cursor is stored on disk, calling consumes it, messages received while nothing was running are not lost, and peek avoids consumption. That is exactly the behavioral context the annotations cannot convey.

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?

Three short sentences, all load-bearing: the core scope first, then the persistence guarantee, then the non-consuming escape hatch. Nothing is redundant and the most important constraint is front-loaded.

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 three-parameter, all-optional read tool with no output schema, the description covers scope, state effects, and the peek escape hatch, which is most of what an agent needs. It omits ordering, whether results drain or repeat, and what a call returns, but these are minor given the schema carries parameter detail.

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 100%, so peek, limit, and includeFromMe are all documented in the schema already. The description's 'without consuming' restates peek's schema text rather than adding syntax or edge-case semantics, so it lands on the baseline 3.

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 pins down a precise resource and scope: 'Messages that have arrived since this tool last ran.' That incremental/new-since-last-run framing clearly separates it from a general search, even though no sibling tool is named. It stops short of a true 5 because differentiation from search_messages and list_conversations is left implicit.

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 only implied: an agent infers it should call inbox to pull new messages rather than search. The one explicit instruction ('call with peek=true to look without consuming') is parameter-level guidance, not when-to-use-this-vs-alternatives guidance. No exclusions or sibling routing are provided.

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