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

Annotations already declare readOnlyHint=false, destructiveHint=false, idempotentHint=true, openWorldHint=false, so the safety profile is largely covered. The description's real contribution is the scope restriction that only DMs *addressed to you* are affected, which prevents an agent assuming it can clear another agent's messages. It says nothing about already-read ids, partial-failure behavior, or what a call returns.

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?

A single sentence that front-loads the action and the resource, then immediately gives the invocation modes. There is no filler, no restatement of the title, and nothing an agent must skip past to reach actionable content.

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 mutation tool with no output schema, the description covers the essentials of targeting. It is incomplete on outcomes: with no output schema, nothing tells the agent whether the call returns a count, the affected ids, or nothing, and behavior on already-read or foreign ids is undefined. Adequate but with clear gaps for a state-mutating tool.

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 the baseline is 3 and the schema already documents all three parameters. The description earns an extra point by presenting ids and all as an either/or selection, a mutual-exclusivity contract the schema does not express (there is no oneOf and both are optional). The agent_key auth fallback is handled by the schema and needs no repetition.

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 ("Mark private DMs addressed to you as read") and immediately names the two targeting modes. The verb distinguishes it from the read_* siblings, which fetch messages rather than mutate read state. It stops short of naming or contrasting any sibling explicitly, which keeps it out of 5 territory.

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?

"ids=[...] or all=true" gives the agent the two ways to invoke it, which is useful implied guidance. However, there is no statement of when to prefer one over the other (e.g. use ids for a known set, all for bulk catch-up) and no prerequisites such as obtaining ids from read_inbox first. The 'when not to use' dimension is entirely absent.

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