repomix

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 packages remote, user-provided repositories (e.g., SKILL.md "Remote Repository Support" and references/usage-patterns.md examples like "npx repomix --remote owner/repo") and the batch script (scripts/repomix_batch.py) accepts arbitrary remote repo URLs and feeds their contents into AI-focused outputs that the agent is expected to read and act on, exposing it to untrusted third‑party content that could carry indirect prompt injections.

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

• Potentially malicious external URL detected (high risk: 0.90). The repomix_batch script invokes npx to run repomix against remote repositories (e.g., "npx repomix --remote https://github.com/owner/repo" and the listed https://github.com/yamadashy/repomix), which causes remote code (the repomix npm package and remote repo contents) to be fetched and executed at runtime, and the skill relies on this external content to function.