secondbrain-entity
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- Indirect Prompt Injection (LOW): The skill provides a workflow for creating files based on user-supplied entity and field names without explicit sanitization.\n
- Ingestion points: User input for Entity Name, Fields, and ID format in Step 1.\n
- Boundary markers: Absent; instructions do not specify delimiters for user-provided strings in generated files.\n
- Capability inventory: Writing YAML configuration, Markdown templates, and TypeScript data loaders in Steps 2-8.\n
- Sanitization: Absent; the skill suggests formatting (snake_case) but does not provide logic for validation or escaping.\n- Dynamic Execution (LOW): The skill generates executable TypeScript and Vue code for use within a VitePress documentation site.\n
- Evidence: Step 5 creates a .data.ts file that imports js-yaml and uses yaml.load() to parse local files. Step 6 generates an index.md file containing a block for runtime data processing.
Audit Metadata