snapshot

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

Beyond the annotations (readOnly, idempotent, non-destructive), the description reveals critical behavior: the return format, frame_ref semantics, live signal state, and the warning that byte-identical frames may be stale/cached and should not be trusted. It also references issue numbers for context, adding depth.

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 moderately long but each sentence contributes: return format, frame_ref usage, signal state, and staleness warning. Minor clutter from issue numbers (#143, #141) slightly distracts but overall is well-organized.

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 compensates by explaining the return tuple and key edge cases. The only notable gap is the undocumented 'profile' parameter, which prevents full completeness for a tool with this richness.

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 schema has one optional 'profile' parameter with no schema description and 0% coverage. The tool description never mentions this parameter, leaving its purpose and possible values completely undocumented. The agent has to guess what 'profile' means.

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 'Capture the current KVM screen (read-only)', which is a clear verb+resource statement. It distinguishes this from siblings by explicitly framing it as a screen capture returning image and JSON, and by contrasting with the 'mouse' tool's use of the returned frame_ref.

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?

It gives explicit guidance on using the returned frame_ref with the mouse tool and instructs to check 'signal' and 'logs' instead when unchanged_since_last_snapshot is true. While it doesn't explicitly state 'use this instead of classify_screen', the alternative actions are clear for the described scenario.

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