Remove photo metadata (EXIF)

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

The description is behaviorally rich — lossless byte-for-byte copying for JPEG/PNG, default retention of an orientation-only EXIF and the ICC profile, applyOrientation forcing a re-encode, and the generated '-clean' filename. However, it describes writing a new file to a path, which conflicts with the readOnlyHint=true annotation implying no environment modification; an agent trusting the annotation could be surprised by output files landing on disk.

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?

Front-loaded with what gets stripped and how, then defaults, then the re-encode caveat — a sensible ordering with no filler sentences. The trailing website URL is the one element that does not directly help an agent invoke the tool, keeping it from a 5.

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 still covers the return artifact (output file named <name>-clean.<ext>) and the returnBase64 path, plus the per-format processing model. It omits what the result payload reports (e.g., which tags were removed or byte savings) and error behavior for unsupported inputs, which are minor.

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 every parameter is already documented, and the baseline is 3. The description adds modest value by tying keepOrientation, keepIcc and applyOrientation to the defaults it applies ('by default the rotation is kept... the ICC colour profile is kept'), but says nothing about the `output` or `image` parameters and does not clarify when `quality` takes effect beyond what the schema states.

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 a precise verb plus an exhaustive resource list ('Remove EXIF, GPS location, XMP, IPTC, comments, PNG text chunks and embedded previews from a photo'), which is far more specific than the title. The natural contrast with the sibling image_exif_read ('remove' vs. 'read') is obvious from the verb, and the per-format handling further pins down what the tool actually does.

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?

'...before sharing it' supplies a clear motivating context, and the format notes implicitly tell the agent when lossless cleaning applies (JPEG/PNG) versus re-encoding (WebP/AVIF/HEIC/TIFF). It never explicitly says when not to use it or names image_exif_read as the alternative when the agent only wants to inspect metadata, so it stops short of a 5.

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