presentation

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

• Third-party content exposure detected (medium risk: 0.30). The skill workflow involves deep web research and code grounding from GitHub/repos via references/research-grounding.md, which ingests external repository and documentation text into the agent's context.

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

• Potentially malicious external URL detected (medium risk: 0.30). The skill references https://sli.dev/llms.txt as a runtime fallback instruction source, which is an unofficial external operational dependency making it a medium tier.