claude-typer
Fail
Audited by Snyk on Nov 21, 2026
Risk Level: CRITICAL
Full Analysis
CRITICAL E004: Prompt injection detected in skill instructions.
- Potential prompt injection detected (medium risk: 0.30). The skill openly advertises rendering a remote composition from a third-party site, which is disclosed behavior, but it also sends the user's prompt to that site and runs its remotely served code without any vetting, and it labels the props with a hidden "MiniMax-M2.7" model value in a Claude-branded tool, which is sketchy but not shown to be malicious.
CRITICAL E005: Suspicious download URL detected in skill instructions.
- Suspicious download URL detected (medium risk: 0.30). The skill directs the agent to fetch and execute third-party code from an unverified personal-looking domain, https://www.laosunwendao.com, which is passed as the Remotion serve URL and loaded as an executable composition bundle, with no pinning or integrity check and no evident link to the official Remotion project, though no concealed download-and-execute chain or decoy source is shown.
MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).
- Potentially malicious external URL detected (high risk: 0.70). The skill runs Remotion CLI against the remote serve URL https://www.laosunwendao.com, loading and rendering a remote composition bundle (executable site code run in headless Chrome) that the skill depends on. The domain belongs to an individual or unknown party with no established first-party or authenticated provenance, so it is weak/unknown and rated high; the npm Remotion packages are pinned official packages with no separate concern.
Issues (3)
E004
CRITICALPrompt injection detected in skill instructions.
E005
CRITICALSuspicious download URL detected in skill instructions.
W012
MEDIUMUnverifiable external dependency detected (runtime URL that controls agent).
Audit Metadata