trump-code-market-signals

Fail

Audited by Gen Agent Trust Hub on Mar 18, 2026

Risk Level: HIGHEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONCREDENTIALS_UNSAFEDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the user to clone an external, untrusted repository from https://github.com/sstklen/trump-code.git and install its dependencies via pip. This allows for the execution of unverified third-party code.
  • [COMMAND_EXECUTION]: The documentation prompts the user to execute multiple local Python scripts (e.g., trump_code_cli.py, chatbot_server.py) and configure an MCP server that runs system commands via python3.
  • [CREDENTIALS_UNSAFE]: The user is requested to manually set sensitive API keys (GEMINI_KEYS, ANTHROPIC_API_KEY, POLYMARKET_API_KEY) as environment variables, which are then accessible to all scripts in the cloned repository.
  • [DATA_EXFILTRATION]: The skill integrates with an external API at https://trumpcode.washinmura.jp for its core functionality, creating a channel for potentially sending sensitive user data or API keys to a server not controlled by the user or a trusted vendor.
  • [PROMPT_INJECTION]: The skill processes untrusted social media posts as primary input data (Ingestion points: realtime_loop.py and data/trump_posts_all.json). It lacks visible boundary markers or sanitization logic to prevent instructions embedded in those posts from influencing the AI's behavior. The system has significant capabilities (Capability inventory: network access via requests, local file system access in data/) that could be exploited via indirect prompt injection.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Mar 18, 2026, 04:58 PM