awesome-ai-security-overview

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

• Third-party content exposure detected (low risk: 0.10). The workflow involves fetching and reading a public README from GitHub (https://raw.githubusercontent.com/gmh5225/awesome-ai-security/refs/heads/main/README.md), which contains community-curated links and descriptions that can be authored by outsiders.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill instructs the agent to fetch the latest curated links from a raw GitHub URL during runtime, but the repository's provenance is unverified and arbitrary relative to the agent's core instructions, representing a weak external runtime dependency.