The Agent Skills Directory

[SAFE]: The skill is a collection of documentation templates and logical rules for organizing research data into chapters. No security threats such as remote code execution, command injection, or data exfiltration were identified.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process external research notes and codebases to generate reports. While this presents an ingestion surface for untrusted data, the provided skill files do not contain exploitable capabilities or dangerous interpolation methods.
Ingestion points: Research notes and codebase paths referenced in SKILL.md and assets/output_templates/report_structure.md.
Boundary markers: Not present in templates; the agent is guided by specific structural patterns instead.
Capability inventory: No file-system write operations, network requests, or subprocess executions are defined in the skill assets.
Sanitization: Not present in the static markdown templates.

chapter-structure