The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes external data that could contain malicious instructions, creating a surface for indirect prompt injection.
Ingestion points: Shell output captured via tmux capture-pane (references/cloud-auth.md) and tool outputs from dbt and databricks CLI.
Boundary markers: No delimiters or safety warnings are used to isolate external data from the agent's instructions.
Capability inventory: The skill uses tmux send-keys to execute commands, performs databricks api post operations, and executes dbt run commands.
Sanitization: The skill lacks explicit sanitization or validation logic for data ingested from shell panes or external logs.
[COMMAND_EXECUTION]: The skill utilizes tmux send-keys to interact with external shell environments. This workflow allows the agent to execute commands using the user's authenticated session, which is necessary for the stated auth tasks but presents a risk if the agent is manipulated into sending unintended commands.
[EXTERNAL_DOWNLOADS]: Instructions include installing the jupyter-databricks-kernel package. This is a vendor-owned resource provided by the skill author to facilitate Databricks-backed notebook execution.

data-platform