meeting

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly fetches and ingests user-generated content from referenced GitLab/GitHub issues (Step 1: "If an issue is referenced ... Fetch the issue details" and Step 3: include issue details and comments as context for sub-agents), which the agent is required to read and use to drive sub-agent behavior and decisions, creating a clear vector for indirect prompt injection.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill explicitly fetches GitHub/GitLab issue content at runtime and injects it into sub-agent prompts (e.g., GitHub/GitLab issue URLs such as https://github.com///issues/123 or https://gitlab.com///-/issues/123), so external content pulled from those URLs would directly control the agent's prompts.