The Agent Skills Directory

[DYNAMIC_EXECUTION]: The skill is designed to run arbitrary Python code provided as input. This code is executed in a remote sandbox environment provided by the vendor.
[INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion and processing of data from external web sources, which introduces a surface for indirect prompt injection.
Ingestion points: The skill uses libraries such as requests, BeautifulSoup, Selenium, and Playwright to fetch content from external URLs as shown in the examples in SKILL.md.
Boundary markers: There are no explicit delimiters or instructions provided to the agent to distinguish between its own logic and instructions that might be present in the scraped data.
Capability inventory: The tool provides a complete Python 3.10 environment with network access and a wide array of powerful libraries for automation and data processing.
Sanitization: The skill does not implement or mention sanitization of the content retrieved from the web before it is processed by the agent.
[EXTERNAL_DOWNLOADS]: The skill references the installation of the belt CLI tool via npx and links to documentation and installation scripts hosted on the vendor's GitHub repository.
[COMMAND_EXECUTION]: The skill utilizes the Bash(belt *) tool to invoke the belt CLI for running Python apps and managing the execution environment.
[REMOTE_CODE_EXECUTION]: Python code execution is performed on a remote infrastructure hosted at inference.sh, which is the vendor's official platform for this service.

python-executor