multi-ai-research

SUSPICIOUS: the skill’s purpose and capabilities mostly align, and the primary installer appears to be the publisher’s official npm/GitHub path. Risk comes from routing research prompts and possibly internal context through OpenCLI/browser-session automation to external AI services, optional transitive skill installation, and ingestion of untrusted external content while retaining write capability. No clear evidence of malware or covert exfiltration, but the trust and data-flow footprint is broader than a simple research helper.