Remove AI watermarks from a directory

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

Annotations declare a non-destructive, non-open-world write, and the description reinforces this by stating results go to a separate directory with a default name of <dir>_clean, so the agent knows originals in the input directory are not overwritten. It does not cover failure handling for unsupported files, overwrite behavior if the output directory already exists, or runtime cost for large batches.

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 sentences, front-loaded with the operation and scope, then the output destination and the mode cross-reference. No filler; every clause carries information the agent needs.

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?

There is no output schema, so the description should say what a batch run reports back and how it treats subdirectories or unsupported files; it says none of that. Given a 4-parameter batch mutation tool, the definition is callable but leaves the caller guessing about recursion, returns, and partial failures.

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 50%, but the description compensates: it spells out media=image vs media=video, routes the mode enum to the remove tool's documented modes, and reveals the output_dir default (<dir>_clean) that the schema omits. The remaining gap is that it does not restate the metadata/visible/all mode meanings inline.

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 verb (clean) plus explicit scope (every supported image or video in a directory, with media switches) makes the batch nature clear, and 'Modes as for the remove tool' signals it is the directory-level counterpart to the sibling 'remove'. It stops short of naming remove/identify as the alternatives an agent should pick between, but the resource and granularity are unambiguous.

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 adds one real precondition — the directory must be owned by the user — and implies batch-vs-single usage by deferring mode semantics to the remove tool. There is no explicit 'use this when you have many files; use remove for one file', so the choice between siblings is left to inference.

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