The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill identifies a potential injection surface by ingesting user-provided data, such as interview notes and customer segments, which are processed to generate business validation scorecards.
Ingestion points: scripts/main.py via the add-interview command's --notes and --segment parameters.
Boundary markers: The instructions in SKILL.md provide structural templates for the agent, though the companion script does not implement programmatic delimiters.
Capability inventory: The skill's capabilities are restricted to reading and writing a local customer_discovery.json file in the execution directory. It does not use network tools, shell execution, or high-privilege system APIs.
Sanitization: Input data is stored and summarized as provided without specific character escaping or filtering.

customer-discovery