claude-tradingview-mcp-autotrading

Warn

Audited by Gen Agent Trust Hub on Sep 16, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPERSISTENCEPRIVILEGE_ESCALATIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the user to download a codebase from a third-party GitHub repository (https://github.com/jackson-video-resources/claude-tradingview-mcp-trading) that is not affiliated with the skill author or a well-known service provider.\n- [REMOTE_CODE_EXECUTION]: The installation process requires running npm install and node bot.js on the cloned external repository, leading to the execution of code from an unverified source.\n- [INDIRECT_PROMPT_INJECTION]: The skill implements a feature to extract trading strategies from YouTube transcripts, creating a vulnerability surface where malicious instructions could be embedded in the transcript data.\n
  • Ingestion points: External YouTube transcript data processed via Apify and provided to the agent as described in the SKILL.md instructions (referencing prompts/01-extract-strategy.md).\n
  • Boundary markers: Absent. The skill lacks explicit delimiters or instructions for the agent to ignore potentially malicious embedded content within the transcripts.\n
  • Capability inventory: The bot script bot.js (as described in SKILL.md) has the capability to execute financial trades on multiple exchanges and write logs to the local file system (trades.csv).\n
  • Sanitization: Absent. There is no evidence of sanitization or validation of the extracted transcript content before it is used to generate trading logic.\n- [PERSISTENCE]: The deployment instructions involve configuring a crontab entry on a VPS to maintain the bot's operation across reboots and scheduled intervals.\n- [PRIVILEGE_ESCALATION]: The VPS setup guide instructs the user to access the server as the root user and use apt to install system-level packages.\n- [COMMAND_EXECUTION]: The workflow involves multiple shell commands for cloning, configuring, and running the application, including git clone, npm install, and direct node execution.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Sep 16, 2026, 10:43 PM
Security Audit — agent-trust-hub — claude-tradingview-mcp-autotrading