get_action_timings

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

No annotations are provided, so the description carries the full burden of behavioral disclosure. It clearly documents default behavior (returns timings for all actions when event_ids is not specified), explains the marker_filter and exclude_markers semantics, and discloses the hardware dependency with 'GPU timing counters may not be available on all hardware/drivers'. The return structure is also fully outlined. Slight lack of explicit note about read-only nature, but this is minor for a query tool.

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?

The description is well-organized with an Args section and Returns section, keeping the purpose statement front-loaded. It is slightly longer than necessary but every section adds value—the note about hardware support is important. Nothing is redundant.

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?

Given the tool's moderate complexity (3 optional params, no output schema provided but described inline), the description covers all necessary details: parameter behavior, return format, and the hardware caveat. The agent has everything needed to call it correctly without additional documentation.

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 description explains each parameter in detail: event_ids defaults to 'all', marker_filter is a partial-match filter, exclude_markers is a list to exclude. It also describes the effect of omitting each one. This goes far beyond the input schema, which only provides types and defaults. The 0% schema coverage means the description fully compensates.

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 opens with a specific verb+resource: 'Get GPU timing information for actions', and explicitly names example actions (draw calls, dispatches). This clearly distinguishes it from sibling tools like get_draw_calls (which lists actions) and get_frame_summary (which summarizes frames).

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 description explains what the tool does and how filters work, but never explicitly contrasts it with alternatives or states when to prefer this tool over get_draw_calls or get_frame_summary. The usage context is implied (for timing data) but no exclusions are given. This leaves the agent to infer selection.

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