vibe-coding-db
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill operates as a high-level router and coordinator. It does not perform sensitive operations like command execution, network requests, or credential handling. Its logic is limited to pattern matching against project configuration files and user input to determine the appropriate database sub-skill to invoke.
- [INDIRECT_PROMPT_INJECTION]: The skill reads external project files to identify the database environment.
- Ingestion points: Reads
package.json,requirements.txt,progress.txt, and database schema files such asprisma/schema.prismato detect database and ORM types. - Boundary markers: Absent; the skill performs direct inspection of file contents to find specific technology strings.
- Capability inventory: None; this coordinator skill only generates routing instructions for the agent and does not possess capabilities for code execution, network access, or sensitive file modifications.
- Sanitization: Absent; the skill relies on exact string matches for identification.
Audit Metadata