The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to explore project context by reading files, documents, and recent commits, which represents a surface for indirect prompt injection.
Ingestion points: Instructions in SKILL.md to read local project files and git history.
Boundary markers: The instructions lack explicit delimiters or safety warnings to ignore instructions embedded within the files being read.
Capability inventory: The skill performs file writes to the .orchestration/ directory and executes local shell commands.
Sanitization: The skill does not mention sanitizing or validating the content retrieved from the project context.
[COMMAND_EXECUTION]: The skill executes a project-relative Node.js script node scripts/orchestration/session.cjs to manage session initialization.

requirement-engineering