tradingview-mcp
Warn
Audited by Gen Agent Trust Hub on Oct 19, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches the
tradingview-mcpserver and its dependencies from a remote GitHub repository (https://github.com/atilaahmettaner/tradingview-mcp) upon its first invocation. - [REMOTE_CODE_EXECUTION]: The server is executed using
uvx, which involves downloading and running remote code. The source repository belongs to an individual user account rather than a recognized trusted organization, which increases the risk profile for code execution. - [INDIRECT_PROMPT_INJECTION]: The skill processes live market data from TradingView and Yahoo Finance. This data enters the agent's context and could potentially contain malicious instructions or adversarial content designed to influence the agent's behavior.
- Ingestion points: TradingView scanner API and Yahoo Finance API (SKILL.md).
- Boundary markers: Absent; there are no instructions to the agent to treat the market data as untrusted or to use specific delimiters.
- Capability inventory: The skill utilizes 39 MCP tools for market analysis and backtesting. While it claims to be read-only, it lacks explicit constraints against processing instructions embedded in the data.
- Sanitization: Not mentioned; the skill does not specify any filtering or sanitization of the incoming market data before it is presented to the agent.
Audit Metadata