TradingView MCP: why it is not safe to use
We found routine third-party data sharing and remote fetches, including an optional hosted endpoint and build-time script install, so we marked this skill warn.
What we found
Our review did not find hidden instructions, hardcoded secrets, privilege escalation, obfuscated code, or tools that perform irreversible actions. The main concern is data flow to third parties during normal use. The skill sends user ticker symbols and query inputs to named external market and news providers so it can return results, including Yahoo Finance, TradingView-backed libraries, optional Marketaux news, and an optional hosted endpoint at `pro.cryptosieve.com`.
We also found remote fetches at runtime from external services, and the Docker build installs `uv` by piping a remote shell script from `astral.sh`. These behaviors are disclosed by the code and appear tied to normal operation, but they still expand the trust boundary beyond the local skill itself.
What to do instead
Use this skill only if you are comfortable sending market queries to those named services. Avoid including sensitive, proprietary, or client-specific information in prompts or ticker-related notes. If you need tighter control, prefer a self-hosted setup, disable optional integrations you do not need, review network egress, and replace the build-time remote script install with a pinned, auditable package installation process where possible.
Source: https://github.com/atilaahmettaner/tradingview-mcp
We report what our security review found at the time we checked, with the goal of keeping people safe. Projects change; if a maintainer has since fixed this, we are glad to recheck it. Email hello@agentpod.com.