The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes untrusted external data which could lead to indirect prompt injection. Ingestion points: scripts/cache.py (via cmd_add and --csv flag) and CLI arguments for name and notes. Boundary markers: Absent. No delimiters are used to wrap stored data. Capability inventory: File system access to read and write contacts.csv. Sanitization: Normalization is applied to identifiers, but no sanitization is performed on descriptive fields.

contact-cache