ralph-update

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

• Third-party content exposure detected (high risk: 0.90). This skill clones and reads files from the public repo https://github.com/Olunuga/ralph-loop (Step 3) and then diffs/copies scripts and "LLM gates" .md files into ralph/ (Steps 4–5), meaning untrusted third‑party content is ingested and can materially influence agent behavior.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill performs a runtime git clone of https://github.com/Olunuga/ralph-loop and copies PROMPT_.md and scripts/gates/llm/.md from that repo into the local ralph/ directory, meaning remote content fetched at runtime can directly control agent prompts and pipeline behavior.