guava-hub

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (low risk: 0.10). The workflow reads third-party skills from arbitrary GitHub repositories referenced in configuration (e.g. SKILL.md line 123-124), leading to low-risk outsider-authored text ingestion.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.70). The skill instructs installation of third-party repositories via npx skills add and claude plugin marketplace update using uncontrolled external locators (romeodiaz/guava-stack, blader/humanizer, mvanhorn/last30days-skill), constituting a weak/unknown/suspicious operational dependency.