feature-domain-expert
Pass
Audited by Gen Agent Trust Hub on Apr 8, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill defines a structured format for business domain documentation, ensuring persistent context is available across development cycles without introducing risks.
- [SAFE]: The data management patterns, including the append-only update discipline for feature files, promote documentation integrity and do not exhibit malicious behavior.
- [SAFE]: References to maintenance scripts like
install.shandsync.shdescribe legitimate project workflows for managing agent environment and memory. - [SAFE]: Indirect Prompt Injection Surface Analysis:
- Ingestion points: Feature documentation files located in
ai-context/features/. - Boundary markers: The skill enforces a specific six-section structure for these files, providing structural context, though no explicit 'ignore instructions' directives are mandated.
- Capability inventory: The agent uses this data as context for proposal and specification tasks and can update it through standard memory management tools.
- Sanitization: No explicit sanitization of content is described beyond the mandated six-section structure.
- [SAFE]: No indicators of prompt injection, data exfiltration, or obfuscation were identified in the skill instructions or provided examples.
Audit Metadata