trump-code-market-signals

Warn

Audited by Gen Agent Trust Hub on Sep 12, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the user to download a full codebase from an untrusted GitHub repository (github.com/sstklen/trump-code.git) which is not associated with the skill's authoring organization.
  • [COMMAND_EXECUTION]: The skill provides explicit instructions to install dependencies via pip and execute various Python scripts (trump_code_cli.py, realtime_loop.py, overnight_search.py, chatbot_server.py) from the downloaded repository. It also configures an MCP server to run a local script (mcp_server.py) downloaded from the same source.
  • [INDIRECT_PROMPT_INJECTION]: The skill has a vulnerability surface for indirect prompt injection by processing external, untrusted social media posts.
  • Ingestion points: Truth Social and X posts are polled every 5 minutes and processed via the realtime_loop.py script and several API endpoints at trumpcode.washinmura.jp.
  • Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the skill's logic description.
  • Capability inventory: The skill uses the requests library for network communication with external domains, reads and writes local JSON data files, and executes shell commands through its MCP server configuration.
  • Sanitization: There is no evidence of sanitization, escaping, or validation of the ingested social media content before it is processed or used by the AI-powered components mentioned in the skill.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Sep 12, 2026, 08:38 PM
Security Audit — agent-trust-hub — trump-code-market-signals