pine-to-python
Pass
Audited by Gen Agent Trust Hub on Sep 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and parse untrusted external data in the form of TradingView PineScript (.pine) files. This creates a surface where malicious instructions hidden in code comments or string literals within the input script could influence the agent's behavior during the translation process.
- Ingestion points: The agent is instructed to parse raw
.pinescripts to identify inputs and logic (SKILL.md). - Boundary markers: The instructions do not define specific delimiters or instructions to ignore natural language commands embedded within the input scripts.
- Capability inventory: The skill generates complex Python code involving
numpy,pandas-ta, andoptuna, which could be used to facilitate malicious actions if the generation logic is subverted. - Sanitization: While the skill includes hardening for financial logic (e.g., repainting and division by zero), it lacks sanitization or validation to prevent the interpolation of instructions from the source script into the agent's operational context.
- [DYNAMIC_EXECUTION]: The skill's primary function is the automated generation of Python scripts from external logic. This involves assembling code components (features, signals, and optimization spaces) based on untrusted input, which is a form of dynamic script generation.
Audit Metadata