integrate-anything
Warn
Audited by Snyk on Jun 28, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).
- Potentially malicious external URL detected (high risk: 1.00). This skill explicitly relies on runtime calls to the Membrane API (https://api.getmembrane.com), which can return clientAction.agentInstructions that the agent is instructed to follow verbatim and also accepts/executed inline JavaScript via the POST /act (--code) path, so external content can both control agent prompts and cause remote code execution.
Issues (1)
W012
MEDIUMUnverifiable external dependency detected (runtime URL that controls agent).
Audit Metadata