The Agent Skills Directory

[PROMPT_INJECTION]: The skill is designed to ingest and process external documents provided by the user, creating a surface for indirect prompt injection attacks.\n
Ingestion points: The skill reads user-supplied syllabi, study notes, and past exam papers from the local environment using the Read, Glob, and Grep tools.\n
Boundary markers: The skill includes a 'Syllabus Guardrail' to ensure content generated is relevant to the provided syllabus, providing a logical constraint, but it lacks specific structural delimiters to isolate external text from internal instructions during processing.\n
Capability inventory: The skill is limited to reading and searching files within the local project environment and does not have capabilities for network communication, external data exfiltration, or arbitrary code execution.\n
Sanitization: There is no evidence of explicit sanitization, escaping, or filtering of the content extracted from the external documents before it is interpolated into the agent's context.

examprep-ai