neo-agent-harness
Pass
Audited by Gen Agent Trust Hub on Jun 15, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is purely instructional and planning-oriented. It does not perform any autonomous file modifications, shell command executions, or network requests. Its primary function is to analyze repository structure and generate a report in Traditional Chinese.
- [SAFE]: No external dependencies, remote scripts, or unpinned packages are referenced or installed. All reference materials are local markdown files provided within the skill package.
- [SAFE]: The skill incorporates strong security and reliability constraints, explicitly instructing the agent to avoid proposing fully autonomous changes for security-critical tasks, compliance, or production deployments.
- [SAFE]: No evidence of prompt injection, obfuscation, or persistence mechanisms was found. The skill encourages best practices such as using secret scanners and deterministic validation tools (linters, type-checkers).
- [SAFE]: The ingestion of repository metadata (e.g., inspecting CI configs and package manifests) is performed solely to assess the 'harnessability' of the environment and does not involve exfiltration of sensitive data.
Audit Metadata