Back to skill

Security audit

Wan 3.0 Prime Reference to Video — Reference-Guided Video on RunComfy

Security checks for vulnerabilities and agentic risk

Overview

The skill matches its video-generation purpose, but it asks users to run an unpinned third-party CLI with account credentials and billing authority.

Install only if you trust the RunComfy CLI and are comfortable sending prompts and approved media URLs to RunComfy under your account. Prefer a pinned, locally installed CLI version, avoid elevated privileges, and confirm cost-sensitive runs before submitting reference videos or high-resolution outputs.

Vulnerability Patterns
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Findings (1)

T08 · Insecure Dependencies

Warning
Location
SKILL.md:71
Finding

Unpinned Third-Party CLI Installation and Execution

Content
View full analysis

Vulnerability Details

File Location: SKILL.md, line 71
Vulnerability Type: Unpinned third-party dependency and unsafe package execution
Risk Level: Medium

Vulnerable Code Snippet

markdown
1. **RunComfy CLI** — `npm i -g @runcomfy/cli` (or `npx -y @runcomfy/cli --version`)

Technical Analysis

The Skill instructs users to install or execute @runcomfy/cli without specifying a reviewed version or verifying package integrity. Both commands resolve a mutable package version from the npm registry:

  • npm i -g @runcomfy/cli installs the package globally and may execute npm lifecycle scripts.
  • npx -y @runcomfy/cli --version automatically downloads and executes the resolved package without interactive confirmation.

Because the package version is not pinned, the code executed by users can change after this Skill has been reviewed. A compromised publisher account, malicious package release, registry compromise, or unexpected upstream change could therefore introduce arbitrary code into the workflow.

The global installation recommendation also unnecessarily broadens persistence and exposure compared with a project-local, version-pinned installation. The CLI implementation is not included in the audited project, so its behavior and the claims made elsewhere in SKILL.md—including endpoint restrictions, token handling, download limits, and absence of telemetry—cannot be independently verified from this artifact.

Attack Path

  1. An attacker compromises the npm publisher account, release pipeline, or another component used to distribute @runcomfy/cli.
  2. The attacker publishes a malicious release under the legitimate package name.
  3. A user follows the Skill instructions and runs either npm i -g @runcomfy/cli or npx -y @runcomfy/cli --version.
  4. npm resolves the unpinned dependency to the malicious release.
  5. Malicious lifecycle scripts or package code execute with the privileges of t ...[truncated 1449 chars]
Remediation
View remediation

Remediation Suggestions

  1. Pin the CLI to an explicitly reviewed version, for example:

    bash
    npx --yes @runcomfy/cli@<reviewed-version> --version
    
  2. Publish and verify the expected npm package integrity hash or signed release provenance before execution.

  3. Prefer a project-local dependency with a committed lockfile over a global installation.

  4. Avoid recommending installation with elevated privileges.

  5. Disable npm lifecycle scripts where compatible with the package:

    bash
    npm install --ignore-scripts --save-exact @runcomfy/cli@<reviewed-version>
    

    If lifecycle scripts are required, document and review them before installation rather than disabling them blindly.

  6. Run the CLI in a constrained environment with access only to the required output directory and RunComfy credential.

  7. Provide a link to the official package repository and documented release-verification process.

  8. Re-audit the pinned CLI whenever the documented version changes.

Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The trigger list includes broad natural-language phrases such as "wan 3.0 prime", "reference to video", and "video from reference images", plus "any explicit ask" wording. Overly broad activation can cause the skill to fire in contexts where the user did not intend this specific local CLI-backed skill, potentially causing unintended external API calls, token use, cost incurrence, or mishandling of user media.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
94% confidence
Finding

The prerequisite suggests running npx -y @runcomfy/cli without pinning a version, which allows whatever package version is current at execution time to run. If the package is compromised upstream or a breaking/malicious release is published, users may execute unexpected code during install/use of the CLI.

Content

No source excerpt is available for this finding.

Vague Triggers

Low
Category
Not specified by scanner
Confidence
88% confidence
Finding

The routing rule says to route to this skill whenever the user said terms like "Wan 3 Prime", "Wan 3.0 Prime", "reference to video" or "ref2v" explicitly, regardless. That unconditional language can override contextual disambiguation and steer requests to this paid, reference-required workflow even when another tool or a clarification step would be safer or more appropriate.

Content

No source excerpt is available for this finding.