The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests user goals and process details to generate new agent skills and scripts. This creates a risk where malicious input could be persisted into new executable artifacts.
Ingestion points: User responses to interview questions in Phases 1, 2, and 3 defined in SKILL.md.
Boundary markers: Absent. The skill does not use delimiters or instructions to ignore embedded commands in the user data.
Capability inventory: The skill writes new SKILL.md files and Python scripts (scripts/utility.py) to the local workspace (Phase 5 of SKILL.md).
Sanitization: Absent. There is no logic to validate or escape user input before it is used to generate new skill files.
[DYNAMIC_EXECUTION]: The skill generates Python scripts and markdown instruction files based on user-defined logic and templates.
Evidence: Phase 5 of SKILL.md and the references/skill-output-template.md describe the automatic creation of a scripts/ directory containing utility.py and a new SKILL.md based on the interview session.

process-interviewer