wan-2-7

Pass

Audited by Gen Agent Trust Hub on Jul 13, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill requires the @runcomfy/cli package, which is the standard tool for interacting with the RunComfy API. It also references a GitHub repository for skill management.
  • [DATA_EXFILTRATION]: Credential management follows best practices, utilizing either the RUNCOMFY_TOKEN environment variable or a configuration file (~/.config/runcomfy/token.json) with restricted permissions (0600).
  • [COMMAND_EXECUTION]: The skill executes the runcomfy CLI to process video generation requests. The documentation explicitly notes that the CLI avoids shell-expansion of user prompts to prevent injection vulnerabilities.
  • [INDIRECT_PROMPT_INJECTION]: The skill allows the inclusion of an audio_url for lip-syncing. As this involves processing external data, it represents a standard indirect injection surface, which the documentation acknowledges as a risk inherent to multi-modal models.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 13, 2026, 04:10 AM
Security Audit — agent-trust-hub — wan-2-7