Filtrix Image Generation

ReviewAudited by ClawScan on May 10, 2026.

Overview

The skill mostly matches its image-generation purpose, but its fal.ai edit path disables a provider safety checker while using API keys and uploading selected images to third-party providers.

Install only if you are comfortable with third-party AI providers receiving your prompts and selected images. Set only the API keys you want the agent to use, monitor API costs, and consider removing or changing the fal.ai edit setting that disables safety checking before use.

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.

Concern

ASI02: Tool Misuse and Exploitation

What this means

fal.ai edits may bypass a provider safeguard and produce or return content the user did not expect to be allowed.

Why it was flagged

The fal.ai edit request uploads the input image and explicitly disables the provider safety checker. That is not necessary for normal image editing and is not called out in the user-facing setup or usage instructions.

Skill content

"enable_safety_checker": False,
        "enhance_prompt_mode": "standard",
        "sync_mode": True,
        "image_urls": [data_uri],

Recommendation

Do not disable provider safety checking by default. Make any safety-related option explicit, documented, and user-controlled.

Note

ASI03: Identity and Privilege Abuse

What this means

Using the skill can consume paid API credits or quota for the configured provider account.

Why it was flagged

The skill uses user-supplied provider credentials. This is expected for image generation, but it gives the skill authority to spend quota or incur charges on whichever provider key is available.

Skill content

| OpenAI | `OPENAI_API_KEY` | platform.openai.com |
| Gemini | `GOOGLE_API_KEY` | aistudio.google.com |
| fal.ai | `FAL_KEY` | fal.ai/dashboard |

Recommendation

Set only the provider keys you intend to use, monitor usage, and ask the agent to confirm provider/model choices for expensive generations.

Note

ASI07: Insecure Inter-Agent Communication

What this means

Sensitive photos or private images provided for editing will be transmitted to an external AI provider.

Why it was flagged

For image editing, the script reads the selected local image, encodes it, and sends it to the chosen provider API. This is purpose-aligned but means local image content leaves the machine.

Skill content

img_b64, mime = _read_image_b64(image)
...
{"inlineData": {"mimeType": mime, "data": img_b64}}

Recommendation

Only edit images you are comfortable sending to the selected provider, and review that provider's data-retention and privacy terms.

Note

ASI04: Agentic Supply Chain Vulnerabilities

What this means

Users have less external provenance information to verify who maintains the code or where updates come from.

Why it was flagged

The registry metadata does not provide a source repository or homepage, while the skill includes runnable scripts. The included code is reviewable and has no package install step, so this is a provenance note rather than a standalone concern.

Skill content

Source: unknown
Homepage: none
...
Code file presence
2 code file(s): scripts/edit.py ... scripts/generate.py

Recommendation

Prefer a published source repository/homepage and complete metadata for credentials and runtime requirements.