🎭 Face Swap — Pro Pack on RunComfy

PassAudited by ClawScan on May 13, 2026.

Overview

This is a coherent RunComfy face-swap skill, but it uses a cloud CLI with a RunComfy token and identity media, so users should confirm consent, privacy, and install provenance before running it.

Install only the official RunComfy CLI, protect your RUNCOMFY_TOKEN, and run this skill only with media and identities you have rights to use. Confirm the model route and inputs before execution, and label or disclose synthetic media when required.

Findings (5)

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.

Low

#ASI02: Tool Misuse and Exploitation

What this means

A run may send selected identity/source media references to RunComfy and create synthetic output, potentially consuming account credits.

Why it was flagged

The skill directs the agent/user to invoke external RunComfy model endpoints with user-supplied media references and write outputs locally.

Skill content

runcomfy run <vendor>/<model>/<endpoint> \
  --input '{"image_url": "...", "identity_url": "..."}' \
  --output-dir ./out

Recommendation

Confirm the selected model, input media, output directory, and expected costs before each run.

Low

#ASI03: Identity and Privilege Abuse

What this means

Anyone or any agent process using this credential can make RunComfy API calls under the user's account.

Why it was flagged

The skill requires a RunComfy credential and local CLI configuration to access the provider account.

Skill content

Required env vars: RUNCOMFY_TOKEN
Required config paths: ~/.config/runcomfy

Recommendation

Use a dedicated or scoped token if available, keep RUNCOMFY_TOKEN private, and revoke it if no longer needed.

Info

#ASI04: Agentic Supply Chain Vulnerabilities

What this means

Installing or running an external CLI can execute code from the package source on the user's machine.

Why it was flagged

The instruction-only skill depends on an external npm-distributed CLI rather than bundled reviewed code.

Skill content

npm i -g @runcomfy/cli      # or:  npx -y @runcomfy/cli --version

Recommendation

Install the CLI from official RunComfy documentation, consider pinning a known version, and avoid running npm/npx commands from untrusted contexts.

Low

#ASI07: Insecure Inter-Agent Communication

What this means

Identity media or media URLs may be processed by RunComfy rather than staying entirely local.

Why it was flagged

The workflow depends on a cloud provider CLI/API for processing face, identity, image, and video data.

Skill content

this RunComfy face swap skill substitutes a face or character into video or still images via the `runcomfy` CLI

Recommendation

Review RunComfy's privacy and retention terms and only provide media you are authorized to upload or reference.

Low

#ASI09: Human-Agent Trust Exploitation

What this means

Misuse could create misleading or non-consensual synthetic media, even though the skill warns against it.

Why it was flagged

The artifact acknowledges that face swapping can be misused and instructs users to confirm rights and refuse harmful requests.

Skill content

**Face-swap is dual-use.** Before invoking any route in this skill, confirm:

- You have rights to the target face

Recommendation

Use only with consent and rights to the media and identity, and disclose synthetic media where required.