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

Annotations already declare a non-destructive, idempotent, closed-world mutation, and the description adds real behavior beyond that: explicit project scoping, that the project identity is returned with the result, that hold_subject freezes the subject cut-out while the background swaps, and strobe flash-safety. It does not say what existing background state is replaced or how the effect is removed (e.g. remove_blur/remove_stylize), so it stops short of full disclosure.

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 dense but front-loaded, opening with the PROJECT-SCOPED constraint before the parameter syntax, and every sentence carries information. The long enumeration-heavy middle sentence is harder to parse than it needs to be, but there is little filler.

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 an 11-parameter mutation tool with 27% schema coverage and no output schema, the description covers nearly every parameter's semantics and notes that the project identity comes back with the result. What is missing is how the effect is undone or how it interacts with overlapping effects, which an agent would likely need.

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 description coverage is only 27%, so the description has to carry the load and largely does: it maps kind values to their meanings and binds colour/colour2 to gradient, style.angle to gradient, style.radius to the blurs, style.hold_f to strobe, source to 'media key | source second | next | previous shot', and hold_subject/hold_t to subject freezing. The sources array and style.stage remain unelaborated, keeping it from a 5.

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 specific verb and resource: 'Replace the background behind the subject over [start, end)'. The enumerated kinds (solid/gradient/blur/radial_blur/source/strobe) and the 'behind the subject' qualifier make the operation more specific than siblings like add_color_screen or blur_region, but no sibling is named to disambiguate explicitly.

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 when-to-use guidance and no alternatives are named (e.g. versus set_composite, add_color_screen, or blur_region), which matters given how many sibling effect tools exist. The only usage-shaped content is a pair of constraints ('Flash-safety applies to strobes', 'No sound, no zoom'), which do not tell the agent when this tool is the right choice.

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