Filtrix AI Image Generation
PassAudited by ClawScan on May 1, 2026.
Overview
This skill coherently generates and edits images through OpenAI, Gemini, and fal.ai, but users should notice that prompts/images and provider API keys are used with third-party services.
Before installing, make sure you are comfortable giving the skill access to the relevant provider API keys and sending prompts or images to OpenAI, Google, or fal.ai. Use only the provider keys you need, watch for API costs, avoid sensitive images unless provider terms are acceptable, and check output paths before writing files.
Findings (4)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
Using the skill may consume credits or bill the user's OpenAI, Google, or fal.ai account.
The skill uses user-supplied provider credentials to call image-generation services. This is expected for the stated purpose, but users should recognize that these keys authorize use of their provider accounts.
Ensure the relevant API key is set as an environment variable: ... OPENAI_API_KEY ... GOOGLE_API_KEY ... FAL_KEY
Set only the API keys for providers you intend to use, keep them private, and monitor provider usage/costs.
Private or sensitive images submitted for editing may be sent to the selected AI provider.
For image editing, the script reads the local input image and embeds it in a request to an external provider. This is necessary for image-to-image editing, but it means image contents leave the local machine.
img_b64, mime = _read_image_b64(image) ... {"inlineData": {"mimeType": mime, "data": img_b64}}Avoid using sensitive images unless you are comfortable with the selected provider's data handling terms.
The skill performs its work by running local scripts that make network requests and write image output files.
The skill instructs the agent/user to run included Python scripts. This is central to the skill's purpose and no automatic install-time execution is evidenced.
python scripts/generate.py --provider <openai|gemini|fal> --prompt "..." ...; python scripts/edit.py --provider <openai|gemini|fal> --image input.png --prompt "edit instruction"
Run the scripts from the reviewed skill package and verify the output path before writing files.
fal.ai edits may rely less on provider-side safety filtering than a user might expect.
The fal.ai edit request disables a provider safety-check option. This is visible in code and limited to the fal edit path, but it is not highlighted in the user-facing setup or provider guide.
"enable_safety_checker": False
If content-safety filtering matters, review or change this setting before using fal.ai editing.