lov-media-crawler
Warn
Audited by Snyk on Nov 17, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (low risk: 0.10). The required workflow probes and downloads a single user-supplied link, and the outsider-authored video title and description returned by the platform's metadata (WeChat Channels feed, or MediaCrawler detail) enter the agent's context through the probe and report output, but the agent only reads content the user explicitly selected.
MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).
- Potentially malicious external URL detected (medium risk: 0.30). The skill clones and runs a pinned commit of the individual-owned MediaCrawler repo (NanmiCoder). It also downloads and executes, with sudo, a release binary from the individual-owned ltaoo/wx_channels_download repo. The binary's SHA256 comes from a checksums file in that same release, so it is not independent authentication. Both are stable, attributable third-party sources with no first-party vendor provenance, so the tier is medium.
MEDIUM W013: Attempt to modify system services in skill instructions.
- Attempt to modify system services in skill instructions detected (medium risk: 0.65). The skill has the agent run
setup-wxclient, which downloads and prepares a third-party binary, and it directs the user to runwxclient.sh start, which requires admin rights, installs a proxy root certificate, and changes the system proxy. The agent is told not to run that step itself, but the workflow still steers the machine toward admin-level security changes, so I flag it under the recall-first rule.
Issues (3)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
W012
MEDIUMUnverifiable external dependency detected (runtime URL that controls agent).
W013
MEDIUMAttempt to modify system services in skill instructions.
Audit Metadata