eve-skill-distillation
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- [General] (SAFE): The skill consists exclusively of instructional text and documentation guidelines. It lacks any executable code, shell scripts, or environment configurations.
- [Indirect Prompt Injection] (SAFE): The workflow identifies a data ingestion surface (recent work, conversation history, reference docs). However, the skill directs the agent to produce structured markdown documentation based on this history rather than executing commands. The provided Authoring Rules (YAML frontmatter constraints, conciseness) serve as structural sanitization.
- [Exfiltration/Exposure] (SAFE): Mentioned paths like '../eve-skillpacks/' and './private-skills/' are for internal project organization and do not target sensitive system locations. No network capabilities are requested.
Audit Metadata