Human notes

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

readOnlyHint=true already covers the safety profile, so the description only needs to add traits. It discloses that results are scoped to live mode and that each note bundles surrounding context (tick, inspect() state, leading frames), which hints at a richer return than a flat list. It says nothing about ordering, limits, or what happens when the session was not live – gaps left entirely unfilled.

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 compact sentence with the resource stated first and the payload contents listed after. Every clause adds information; nothing is padded or repeated from the title.

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?

With no output schema, the description must carry the return-value burden, and it does: it names the human's words, the tick, the inspect() state at that moment, and the preceding frames. The live-mode constraint is stated. What is missing is ordering/pagination behavior and whether prior game capture is a precondition.

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?

The tool takes zero parameters, so the schema carries no semantics to supplement; the baseline for a parameterless tool is 4. The description correctly introduces no parameter discussion, avoiding noise.

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 identifies a specific resource – moments the human flagged in the overlay – and enumerates the payload (their words, the tick, the inspect() state, preceding frames). It is distinctive enough that an agent won't confuse it with game_state or game_frames, though it never states an explicit verb and never names a sibling to differentiate against.

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?

The parenthetical '(live mode)' implies the operating condition under which the tool applies, which is genuine usage context. However, there is no explicit when-to-use guidance, no when-not-to-use, and no routing to alternatives such as game_frames or game_capture for the same underlying data.

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