qodo-pr-resolver

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

• Third-party content exposure detected (high risk: 0.85). At runtime the skill fetches Qodo PR/MR comments (including “Code Review by Qodo” bodies and inline review comment text) from the git provider via CLI/API (e.g., gh pr view ... --json comments and gh api .../pulls/<pr-number>/comments), where the comment text is outsider-authored by Qodo bots and then ingested into the agent context for parsing and prompt-following.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill fetches Qodo review comments at runtime from provider APIs (e.g., GitHub: gh api repos/{owner}/{repo}/pulls//comments) and then executes the fetched Qodo "agent prompts" as direct instructions, so the external API content directly controls the agent's actions.