The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data which could contain malicious instructions designed to influence the agent's behavior during tuning or evaluation.
Ingestion points: The skill ingests data from user-provided query files (via the --queries flag) and retrieves data samples or search results from the Viking AI Search application dataset.
Boundary markers: The instructions do not define explicit boundary markers or provide the agent with instructions to ignore embedded commands within the ingested data.
Capability inventory: The agent has the capability to execute vs CLI commands that perform network requests, write diagnostic artifacts and status to the local .viking/search-tuning directory, and apply new search configurations through the apply command.
Sanitization: There is no requirement in the skill instructions to sanitize or validate the content of the external query files or dataset samples before they are passed to LLM-based commands for generation or relevance judging.

vs-search-tuning