The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests and updates personnel profiles and management context from local files, which could serve as an injection surface if these files contain untrusted or malicious instructions. * Ingestion points: Reads and writes to .agents/em-context.md and .agents/reports/[name].md to store and retrieve situational context. * Boundary markers: There are no explicit delimiters or instructions provided to the agent to treat content within these files as untrusted data or to ignore embedded instructions. * Capability inventory: The skill uses file-read and file-write capabilities to manage personnel records and management profiles. * Sanitization: No content validation or sanitization routines are specified for the data stored in or retrieved from the context files.

difficult-situations