update-skills
Warn
Audited by Snyk on Nov 19, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (medium risk: 0.30). The required step 7 files or finds the drift issue by reading existing issues in the public Gharib89/skills repo (outsider-openable, so their titles and text reach the agent), and step 8 also reads upstream commit subjects from third-party repos.
MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).
- Potentially malicious external URL detected (medium risk: 0.30). Step 2 runs
npx skills@1.8.1 add Gharib89/skills --skill …, which installs skill material from a remote repository that the agent then relies on. The publisher is an individual-owned repository, so provenance is common/unofficial, which makes this medium; the pinned composed installs use the same mechanism, but the pins and sources are not shown to be first-party, so they do not raise the tier.
Issues (2)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
W012
MEDIUMUnverifiable external dependency detected (runtime URL that controls agent).
Audit Metadata