redact_region

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

No annotations are provided, so the description carries the full burden: it discloses that the default box style is irreversible, that blur/mosaic are partially recoverable, and that out-of-frame regions are rejected rather than silently no-ops. That is exactly the safety-relevant context an agent needs for a destructive edit. It omits what happens to concurrent defaults and what the call returns, costing it a point.

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?

Four short sentences, front-loaded with the scope, then the caveat, then the destructive-default warning, then the failure behavior. Nothing is repeated and every sentence carries decision-relevant information.

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 7-parameter mutation tool with no annotations and no output schema, the description covers scope, destructiveness, and error behavior well. The remaining gap is the return value — with default output landing under SCREENCAST_HOME/edits, the agent cannot tell from the description whether the resulting path is returned or must be inferred.

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 100%, so parameter meaning is already established; the description still adds value by explaining why the style choice matters (recoverability), which the schema only asserts briefly. It also reinforces the failure mode for bad region values. Minor added value over the schema, hence short of 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?

States a specific verb and resource ("Redact declared rectangles in a video") and immediately scopes it against the adjacent capability it is not: "it is NOT automatic secret detection." An agent can distinguish this from OCR-driven siblings like find_text without opening the schema.

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?

Gives a clear when-to-use condition (you already know the rectangles) and an explicit when-not (not automatic secret detection), which steers the agent away from using it as a detector. It stops short of naming the alternative route (e.g. find_text to locate secrets first), so the guidance is clear but not fully routed.

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