The Agent Skills Directory

[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by ingesting metadata from local agent files and interpolating that data into project documentation.
Ingestion points: Reads YAML frontmatter from .claude/agents/*.md and content from .claude/rules/team.md as described in Phase 1 and Phase 4 of SKILL.md.
Boundary markers: Absent. No explicit delimiters or instructions are used to prevent the agent from obeying instructions embedded in the scanned files.
Capability inventory: The skill uses Read, Write, and Edit tools to modify files within the project's .claude/ configuration directory.
Sanitization: Absent. File content and metadata are used directly in templates without validation or escaping.
[SAFE]: The skill references official documentation for Claude Code agents on Anthropic's established domain.

clnode-agents