The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and act upon instructions found in local project files, which constitutes a potential injection surface.
Ingestion points: Files identified via patterns like .cursor/rules/*.mdc, CLAUDE.md, AGENTS.md, and other documentation discovered via Glob and Grep tools in SKILL.md.
Boundary markers: Absent. The skill does not define specific delimiters or instructions to ignore embedded commands within the rule files it reads.
Capability inventory: Uses Glob, Grep, and Read tools. The skill instructs the agent to analyze and prioritize rules based on enforcement levels (CRITICAL, MANDATORY) found in the content.
Sanitization: Absent. Data from external files is processed and categorized without explicit validation or escaping of the instructional content.

find-rules