Metadata MCP Connector

Fix Flexible Creative Image

fix_flexible_creative

Create an AI-edited copy of a supplied background or flat creative image, at the source's pixel size (longest edge at most 2000 px) when the shape is close to the source's. Returns a hosted URL for a new standalone image that the caller can download or use independently. It does not update the source, a library item, an ad, or the layered document. The first-party editor can use the returned URL as a replacement background. Registered brand characters shown in the source are kept unless characters is passed to change them. WARNING: text baked into a flat image may be repainted or become unreadable; separate text and logo layers are outside this tool's edit.

Input Schema

Name	Required	Description
`domain`	Yes	Company domain for brand context (e.g. 'metadata.io').
`characters`	No	The brand characters the edit must keep. Omit to keep the ones the image shows; [] drops them all, a list keeps only those names.
`image_source`	Yes	URL of the background image to edit (PNG/JPG/WebP). Typically the flexible creative's current background.
`edit_instruction`	Yes	Natural-language description of the fix(es). Can combine several changes in one request.

Schema Changelog

Changes observed during successful MCP inspections.

Changed1 schema field changedOct 21, 2026

addedInput schema / properties / characters

Added value: +{
+  "description": "The brand characters the edit must keep. Omit to keep the ones the image shows; [] drops them all, a list keeps only those names.",
+  "items": {
+    "type": "string"
+  },
+  "maxItems": 5,
+  "type": "array"
+}

First observedSep 16, 2026

TDQS

A4.1/5.0

Behavior5/5

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

Goes well beyond the annotations (readOnly=false, destructive=false, openWorld=true): it discloses no side effects on source/library/ad/layered document, the output form (hosted URL), the resolution cap (longest edge ≤2000 px), character-preservation defaults, and a real hazard warning that baked-in text may be repainted or become unreadable.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Front-loads the action and the output form, then moves to constraints and warnings. It's efficiently written, though the sentence about the returned URL being downloadable/independent and the follow-on sentence about the first-party editor reusing it as a replacement background partially overlap.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

No output schema exists, and the description carries that burden by specifying the return value (a hosted URL to a new standalone image the caller can download). Combined with the size cap, no-side-effect guarantees, and the text-baking warning, an agent has everything needed to call it and interpret the result.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so all four parameters are already documented, so the baseline is 3. The description adds only the notion of 'registered brand characters' being kept unless characters is passed, which slightly enriches the characters parameter but largely restates the schema's own 'Omit to keep the ones the image shows' wording.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb+resource: 'Create an AI-edited copy of a supplied background or flat creative image', and scopes the output ('hosted URL for a new standalone image'). It implicitly separates itself from sibling edit paths by saying it 'does not update the source, a library item, an ad, or the layered document', but never names the sibling to use instead (e.g. edit_brand_creative).

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives a clear context of use: when you want an independent edited copy at roughly the source's size, with the returned URL usable as a replacement background in the first-party editor. The exclusions ('does not update the source, a library item, an ad, or the layered document') tell an agent when not to expect side effects, but no explicit alternative tool is named for those cases.

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

All Metadata MCP Connector tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources

Need Help?