kud-chart-author
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: No malicious injection patterns such as 'ignore previous instructions' or 'reveal system prompt' were detected. The skill uses instructional language consistent with its' purpose as a curriculum authoring tool.
- [DATA_EXFILTRATION]: No commands for network access (curl, wget) or access to sensitive file paths (~/.ssh, .env) were found. The skill does not handle sensitive credentials.
- [REMOTE_CODE_EXECUTION]: The skill does not download or execute any external scripts or binaries. It relies purely on the model's internal reasoning to generate curriculum content.
- [OBFUSCATION]: No obfuscated content, Base64 strings, or hidden Unicode characters were detected. The Markdown and YAML contents are in plain text.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data via the 'primary_input' field (e.g., raw curriculum documents).
- Ingestion points: The 'primary_input' field in SKILL.md.
- Boundary markers: Absent (uses simple interpolation placeholders).
- Capability inventory: Limited to text generation; no subprocess execution, network calls, or file-writing capabilities.
- Sanitization: Not present.
- Due to the lack of sensitive capabilities (shell/network), this represents a standard text-to-text transformation surface with no significant security risk.
- [DYNAMIC_CONTEXT_INJECTION]: No use of the backtick-command syntax (!
command) was found in the SKILL.md file.
Audit Metadata