The Agent Skills Directory

[DYNAMIC_EXECUTION]: The skill is designed to execute arbitrary Python code provided via the 'code' parameter using the 'belt' CLI tool, which is the primary intended functionality.
[INDIRECT_PROMPT_INJECTION]: The skill provides tools for web scraping and data processing, which creates a surface for ingesting untrusted content from external sources. 1. Ingestion points: The code parameter allows fetching external data using libraries like requests or selenium. 2. Boundary markers: No specific delimiters or safety instructions are defined in the schema for the external content being processed. 3. Capability inventory: The environment provides a full Python 3.10 interpreter with network and file access within a sandbox. 4. Sanitization: The skill relies on the isolation and sandboxing provided by the inference.sh platform.
[EXTERNAL_DOWNLOADS]: The skill references installation scripts and documentation from the official inference-sh GitHub repository and the inference.sh service domain. These are well-known resources associated with the execution platform.

python-executor