The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process customer data such as usage metrics and support history, which represents a potential surface for indirect prompt injection if that data contains malicious instructions. * Ingestion points: Customer usage metrics, support history, and NPS/CSAT trends referenced in SKILL.md. * Boundary markers: None. * Capability inventory: File system writes for templates and execution of a local Python script for validation. * Sanitization: None.
[COMMAND_EXECUTION]: The skill includes a local script, scripts/check-output.py, which the agent uses to validate that outputs contain required sections. This is a benign, local operation using standard Python libraries.

qbr-planning