wiki-dashboard
Pass
Audited by Gen Agent Trust Hub on May 19, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill performs local wiki analysis and visualization using provided Python scripts. It adheres to strict design constraints, including no network access and no external dependencies.
- [COMMAND_EXECUTION]: Uses Bash to execute the provided Python scripts (
render_dashboard.py,build_graph.py) to process wiki data and generate HTML reports. This execution is limited to the skill's internal logic and is consistent with its stated purpose. - [DATA_EXFILTRATION]: All data processing is performed locally on the user's machine. The generated dashboards are self-contained HTML files saved to the local wiki directory, and the skill implementation contains no network-capable code.
- [PROMPT_INJECTION]: The skill facilitates an LLM evaluation loop to identify semantic relationships between wiki pages. It uses structured prompt templates and sanitizes data via HTML escaping in the generated reports, following best practices to mitigate indirect injection risks.
Audit Metadata