skill-pipeline
Pass
Audited by Gen Agent Trust Hub on Mar 31, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill operates as a metadata-driven orchestrator and does not contain any executable scripts, binary files, or automated shell commands.
- [SAFE]: It interacts solely with project-specific documentation and plan files, such as plan files in the docs directory and local configuration files like CLAUDE.md. It does not attempt to access sensitive system directories, SSH keys, or cloud provider credentials.
- [SAFE]: The orchestrator incorporates human-in-the-loop validation, such as requiring plan approval before beginning the automated implementation phase.
- [SAFE]: No obfuscation, data exfiltration patterns, persistence mechanisms, or remote dependencies were identified within the skill or its reference documentation.
Audit Metadata