parallel-research

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

• Third-party content exposure detected (high risk: 1.00). The skill's run-agent.sh and agent prompts explicitly instruct spawned agents to perform WebSearch/WebFetch ("--allowedTools 'WebSearch,WebFetch,...'") and "search the web" (Phase 3 prompt.md and Phase 6 launcher), so it fetches and ingests untrusted public web content which the agents then read and use to drive research outputs and synthesis.