tauri-app-sql
Pass
Audited by Gen Agent Trust Hub on Jun 23, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: References official Tauri documentation (v2.tauri.app) for SQL plugin setup and usage, which are well-known technology service resources.
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface because it generates database configurations and query code based on user-provided descriptions of their database needs.
- Ingestion points: User requests for SQL integration, schema design, or query logic defined in SKILL.md.
- Boundary markers: Not present; the instructions do not define delimiters or specific warnings for processing user-provided data.
- Capability inventory: Database access via
tauri-plugin-sqland capability definitions intemplates/capabilities.json(providing access tosqlite:test.db). - Sanitization: Not present; the skill does not instruct the agent to escape or validate user-provided input before using it in generated SQL templates.
- [SAFE]: The
templates/capabilities.jsonfile uses broad wildcards for SQL permissions (e.g.,INSERT INTO *andSELECT * FROM *). While this represents a deviation from the principle of least privilege, it is a standard pattern in development templates and is restricted to the specific database resourcesqlite:test.db.
Audit Metadata