github-issue-fix

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly instructs the agent to fetch and ingest GitHub issue content using gh issue view (SKILL.md PLAN step 1), which is user-generated, third-party content that the agent must read and act on, so it could carry indirect prompt-injection instructions.

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

• Potentially malicious external URL detected (high risk: 0.90). This skill uses the GitHub CLI to fetch live GitHub issue content at runtime (e.g., https://github.com///issues/), and that fetched issue text is incorporated into and directly controls the agent's planning/instructions, so external URLs to GitHub issues are a runtime dependency that can control prompts.