boss-job-search

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill installs the wuying-agentbay-sdk package from PyPI, which is the official software development kit for the AgentBay platform.
  • [COMMAND_EXECUTION]: The skill utilizes a Python script scripts/browser-use.py that executes a remote browser task. The task instructions are passed as a command-line argument and processed by the AgentBay SDK to control a browser session.
  • [INDIRECT_PROMPT_INJECTION]: The skill has a potential surface for indirect prompt injection because it incorporates user-supplied search parameters into natural language instructions for the browser agent.
  • Ingestion points: The task argument in scripts/browser-use.py, which is constructed from user input to define the browser's actions.
  • Boundary markers: No explicit delimiters or boundary markers are used in the instruction strings to separate user data from the browser control logic.
  • Capability inventory: The remote browser agent has capabilities including web navigation, interaction with page elements (clicking, typing), and visual observation of the page state.
  • Sanitization: The skill does not perform explicit sanitization or escaping of the user input before it is interpolated into the browser task description.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 12:30 PM