crossframe-dialogue
Pass
Audited by Gen Agent Trust Hub on Aug 10, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is entirely instructional and documentation-oriented. It contains Markdown templates and protocols for dialogue management in Chinese, with no executable scripts, binaries, or shell commands present in the analyzed files.
- [SAFE]: All referenced dependencies and routing instructions use local relative paths (e.g.,
../crossframe/SKILL.md), indicating a modular internal structure rather than reliance on external or untrusted remote resources. - [SAFE]: The skill incorporates explicit safety protocols (e.g.,
consultation-boundary-protocol.md) that instruct the agent to avoid providing legal, medical, or psychological advice and to downgrade judgement levels when evidence is insufficient. - [SAFE]: No obfuscation, hardcoded credentials, or persistence mechanisms were found. The use of YAML metadata and Markdown documentation follows standard best practices for agent skill development.
Audit Metadata