browser-tape-interceptor

Warn

Audited by Gen Agent Trust Hub on Oct 14, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONDATA_EXFILTRATIONPRIVILEGE_ESCALATIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructions specify connecting to a Chrome instance using the --remote-debugging-port=9222 flag. This allows for low-level interaction and command execution within the browser environment via the Chrome DevTools Protocol (CDP).
  • [DATA_EXFILTRATION]: The primary purpose of the skill is to intercept (menyadap) WebSocket data frames from financial trading platforms. This includes sensitive information such as transaction times, prices, and volume. The ability to listen to Network.webSocketFrameReceived within an authenticated session creates a significant risk for data exposure if the agent is instructed to forward this data externally.
  • [PRIVILEGE_ESCALATION]: By attaching to a browser session where a user is already logged in, the agent gains access to the security context of the user's trading account without needing to perform its own authentication. This level of access typically bypasses standard application boundaries.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests live data streams from external web brokers. Because the agent processes these external payloads (e.g., parsing WebSocket frames), there is a potential surface for indirect prompt injection if the ingested data contains instructions that could manipulate the agent's subsequent logic or outputs.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Oct 14, 2026, 03:11 AM