scraping-specialist

Fail

Audited by Gen Agent Trust Hub on Apr 5, 2026

Risk Level: HIGHEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill's setup process fetches the uv package manager from astral.sh and clones the ida-domain repository from the HexRaysSA GitHub organization. These sources are recognized as well-known service providers and official vendors for reverse engineering software.
  • [COMMAND_EXECUTION]: Several utility scripts (setup.py, run.py, setup.sh) utilize subprocess.run or shell commands to manage environment dependencies, including git, npm, and uv. This is necessary for bootstrapping the scraping and analysis environments.
  • [REMOTE_CODE_EXECUTION]: The binary analysis component programmatically writes Python scripts to temporary files in /tmp and executes them using the uv runtime. This design allows for flexible automation of IDA Pro tasks based on agent-generated logic.
  • [PROMPT_INJECTION]: The skill ingests untrusted data from external sources, including binary file contents and web page HTML (X/Twitter, Trustpilot, Google Ads). This creates a surface for indirect prompt injection where malicious content in the data could attempt to influence the AI agent's reasoning or subsequent actions.
Recommendations
  • HIGH: Downloads and executes remote code from: https://astral.sh/uv/install.sh - DO NOT USE without thorough review
Audit Metadata
Risk Level
HIGH
Analyzed
Apr 5, 2026, 08:04 AM