dashpilot

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The toolset explicitly returns captured HTTP requests, responses and cached query data (which may include Authorization headers, cookies, API keys or passwords) and the workflows instruct inspecting those responses, so the agent will likely receive and may need to surface secret values verbatim.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). This extension + MCP server explicitly captures network requests/response bodies, WebSocket frames, React/query-cache data and exposes them via native messaging to an external agent/LLM, enabling intentional data exfiltration and credential/token theft.

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

• Third-party content exposure detected (high risk: 0.85). The required workflow uses DashPilot MCP tools like page_snapshot, get_network_log, and get_websocket_log to capture and return HTTP response bodies and page/React-derived text from the active browser tab, which can include outsider-authored free text from arbitrary web pages the user navigates to at runtime.