The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill provides a mechanism to read terminal screen text, which could contain untrusted data or malicious instructions. * Ingestion points: The superset terminals read command in SKILL.md allows the agent to ingest the contents of active terminal panes. * Boundary markers: The skill lacks explicit delimiters or instructions to ignore or isolate instructions found within terminal output. * Capability inventory: The skill possesses high capabilities, including the execution of arbitrary shell commands in new panes via the superset panes split and new-tab commands. * Sanitization: No filtering or sanitization of terminal content is performed before it is provided to the agent.
[COMMAND_EXECUTION]: The skill is primarily designed to execute shell commands in terminal panes using the --command argument of the superset panes CLI tool.
[EXTERNAL_DOWNLOADS]: The skill utilizes the superset update command to fetch and install updates for the Superset CLI tool from the vendor's infrastructure.
[DYNAMIC_EXECUTION]: The skill facilitates the dynamic generation and execution of shell command strings within terminal environments at runtime.

panes