darwinian-evolver

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

• Third-party content exposure detected (medium risk: 0.30). The runtime path is scripts/parrot_openrouter.py: user-provided prompt_template/artifact (including {{ phrase }} rendering) and the generated failure cases are interpolated into LLM prompts and sent to the provider via _prompt_llm()/chat.completions.create, so outsider text supplied through the skill inputs can be ingested without selecting a specific trusted item first.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill's install step explicitly clones and later runs code from the upstream repo (git clone --depth 1 https://github.com/imbue-ai/darwinian_evolver.git), so remote code is fetched and executed as a required dependency.