btc-momentum-analyzer

Warn

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • COMMAND_EXECUTION (MEDIUM): The scripts quick_test.sh and test.sh execute system commands to run Python processes and pass data between them, which could be exploited if input parameters are not properly validated.
  • DYNAMIC_EXECUTION (MEDIUM): Use of Python 'here-docs' (python3 << 'EOF') in quick_test.sh and string execution (python3 -c) in test.sh involves running code generated or embedded at runtime. This pattern is used for dynamic logic assembly which can bypass static security analysis.
  • EXTERNAL_DOWNLOADS (LOW): The skill is designed to interact with the OKX API (via the referenced fetch_btc_data.py) to download market data. While targeting a legitimate service, this creates an external dependency and an entry point for untrusted data.
  • DATA_EXPOSURE (LOW): Multiple files (README.md, QUICKSTART.md, quick_test.sh, test.sh) contain hardcoded absolute paths (/Users/adrian/Desktop/BA/MACD/). This exposes the local file system structure of the author/user and limits the portability of the skill.
  • INDIRECT_PROMPT_INJECTION (LOW): The skill processes external financial data to generate trading advice, creating a surface for indirect injection.
  • Ingestion points: Market data is fetched from the OKX API through the scripts/fetch_btc_data.py script.
  • Boundary markers: No boundary markers or specific 'ignore embedded instructions' warnings are present in the provided scripts.
  • Capability inventory: The skill can execute shell scripts, run Python code, and perform network requests.
  • Sanitization: No explicit sanitization or validation logic for the external API response data is present in the provided script files.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 17, 2026, 05:56 PM