research-lookup

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

• Third-party content exposure detected (high risk: 0.85). At runtime, the skill calls external web/academic search backends (e.g., Perplexity via OpenRouter and Parallel Chat API) which return synthesized text and excerpts from public web pages/scholarly sources; that returned free-form content is then inserted into the agent’s LLM context as response (and related sources/snippets) in research_lookup.py → lookup() → format_response().

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

• Potentially malicious external URL detected (high risk: 0.90). The README explicitly instructs running "curl -fsSL https://parallel.ai/install.sh | bash" to install the required parallel-cli (a required dependency), which fetches and executes remote code during setup/runtime (https://parallel.ai/install.sh).