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

Annotations already declare idempotentHint=true, and the description reinforces this with 'Calling it again edits the settings rather than stacking.' It adds meaningful behavioral details: it is project-scoped, applies to the whole picture (captions and overlays scale), changes no timing, and returns the project identity. These go beyond the annotation hints, though it doesn't discuss side effects like permissions or error handling.

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 lengthy but each sentence adds value: scope, visual intent, parameter ranges, behavior on re-invocation, effect on overlays, and removal. It is front-loaded with the project-scoped warning and the core purpose, and the parameter details are logically grouped. Slightly overlong, but justified given the parameter count and complexity.

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 7 parameters and no output schema, the description covers the essential context: what it does, parameter semantics, idempotent behavior, scope, and relationship to remove_screen_frame. It even mentions the return value ('returns the project identity'). It doesn't cover error conditions or prerequisite conditions beyond the project_id, but given the available information, it is substantially complete.

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 only 14% (only project_id has a description), so the description carries the full burden for the other six parameters. It provides concrete ranges (inset 0.02-0.35 with default 0.08, radius 0-0.25 as a fraction, shadow 0-1), color format (#RRGGBB), and the direction enum options. This is far more than the schema provides, effectively compensating for the coverage gap.

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 'PROJECT-SCOPED' and then defines the exact visual treatment: 'the standard treatment for a screen recording or app demo' with inset, rounded corners, shadow, and gradient backdrop. It distinguishes itself from generic frame tools by naming the specific aesthetic, and even ties its gradient renderer to add_color_screen, providing a clear identity among siblings.

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 explicitly frames itself as 'THE tool for' that look, indicating when to use it. It also explains that calling again edits rather than stacks, and how to remove it with remove_screen_frame. However, it doesn't explicitly contrast with similar tools like set_frame or add_screen_takeover, so the exclusion guidance is partial.

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