The Agent Skills Directory

[SAFE]: The skill is intended for domain modeling and documentation management. No malicious behaviors such as credential theft or remote code execution were found.
[PROMPT_INJECTION]: No prompt injection patterns were detected. The instructions focus on canonical terminology and architectural decisions.
[DATA_EXFILTRATION]: No exfiltration patterns were detected. The skill only interacts with local project documentation files.
[EXTERNAL_DOWNLOADS]: No external downloads or remote dependencies are utilized.
[PROMPT_INJECTION]: Analysis of indirect prompt injection potential (Category 8): 1. Ingestion points: User-provided domain terms and scenarios in the conversation. 2. Boundary markers: Markdown templates define the structure for stored data. 3. Capability inventory: Reading and writing to .md files (CONTEXT.md, ADRs) within the project directory. 4. Sanitization: Instructions to challenge fuzzy language and cross-reference with code provide a semantic layer of validation.

domain-modeling