The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill provides instructions to install the deepline CLI globally using npm install -g deepline. It also refers users to an external installation guide at https://code.deepline.com/INSTALL.md.
[COMMAND_EXECUTION]: The skill utilizes the deepline CLI for operations such as deepline auth status and deepline tools execute. It also targets a local runtime binary at <workspace-root>/.deepline/runtime/bin/deepline if the global command is unavailable.
[INDIRECT_PROMPT_INJECTION]: The skill acts as a bridge to external data sources including Snowflake tables (e.g., AERO_DB.CONVEX_RAW.*), Axiom production logs, and RudderStack events.
Ingestion points: Data is ingested through tool calls like snowflake_run_semantic_query and snowflake_run_query which fetch results from external warehouses and logging services.
Boundary markers: The skill relies on structured JSON payloads for queries but does not explicitly define prompt-level delimiters to isolate ingested data from agent instructions.
Capability inventory: The agent can execute CLI commands and perform warehouse queries via vendor-provided tools.
Sanitization: No specific sanitization or filtering logic is described for the data returned from external sources before it is processed by the agent.

deepline-analytics