memory_stats

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

Annotations already declare readOnlyHint, idempotentHint, and destructiveHint=false, so the safety profile is covered and the description's 'Read-only' adds nothing new. What it does add is the content scope: which memory categories are reported. It stops short of saying whether figures are live or sampled, or what the top-texture ranking is based on, so it is adequate rather than rich.

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?

Two short fragments, heavily front-loaded with the classification tag and scope of output. The enumeration is terse and functional. Minor waste in the '[INSPECT] Read-only' prefix, which restates what the annotations already declare.

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 zero-parameter inspection tool with no output schema, the description does the necessary work by telling the agent exactly which memory categories come back. The only missing element is how these stats relate to the other perf/geometry tools, which the description does not address.

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 there is no parameter semantics to convey and the baseline of 4 applies. The schema is an empty object with full coverage, and nothing in the description misleads about inputs.

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 states a concrete resource — GPU memory — and enumerates its breakdown (geometry buffers, texture VRAM, instance data, top textures by size). That separates it from siblings like geometry_details or texture_details, which drill into a single asset. It is clear, though the sentence is a noun-phrase enumeration rather than an explicit verb+object statement, so it falls just short of the top score.

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?

There is no guidance on when to reach for memory_stats versus performance_snapshot, draw_call_breakdown, or perf_monitor, all of which live in the same performance-adjacent cluster. The '[INSPECT]' tag implies a read-only inspection class but nothing states the condition that selects this tool over its neighbours.

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