codex-issue-runner

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

• Third-party content exposure detected (high risk: 0.90). This skill explicitly instructs the agent to run commands like gh issue view N and gh pr view N (see Step 4 and "Task Formatting Tips" in SKILL.md) so the agent reads GitHub issues/PRs — untrusted, user-generated third‑party content — and then uses that content to drive coding decisions and actions.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill explicitly tells the agent to run "gh issue view N" to fetch and read GitHub issue content (e.g. https://github.com///issues/N or the API https://api.github.com/repos///issues/N) at runtime, which injects external content directly into the agent's prompt and is required for the workflow.