The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill creates a surface for indirect prompt injection by fetching data from external web sources and processing it within an environment that has code execution capabilities.
Ingestion points: The skill retrieves untrusted data from the internet using search tools such as tavily/search-assistant, exa/search, and exa/answer via the belt CLI.
Boundary markers: There are no explicit markers or instructions defined to isolate the content retrieved from the web from the rest of the agent's context.
Capability inventory: The skill has access to the Bash tool and uses the infsh/python-executor application to execute Python scripts at runtime.
Sanitization: No mechanisms for filtering or sanitizing the retrieved search results are implemented before the data is utilized in prompt generation or script execution.
[DYNAMIC_EXECUTION]: The skill generates and runs Python code during runtime to create visual charts, using the infsh/python-executor tool within the belt environment.
[EXTERNAL_DOWNLOADS]: The skill instructs users to install a CLI tool from belt-sh/cli and references an external installation script hosted at https://raw.githubusercontent.com/inference-sh/skills/refs/heads/main/cli-install.md.

case-study-writing