oma-backend
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill enforces secure coding standards, including the use of parameterized queries and ORM-based data access to prevent SQL injection attacks.
- [SAFE]: Credential management instructions explicitly forbid hardcoding secrets, directing the agent to use environment variables or secret managers for database URLs, API keys, and other sensitive configuration.
- [SAFE]: Dependency management uses standard package registries and tools (npm, pip, cargo) without suspicious remote execution patterns or untrusted source downloads.
- [INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it is designed to read codebase files, manifests, and API contracts to determine the tech stack.
- Ingestion points: Project manifests (package.json, pyproject.toml), backend source code, and database migrations (SKILL.md).
- Boundary markers: The instructions do not define specific delimiters for reading project files, although they mandate validation of all inputs.
- Capability inventory: The skill has the ability to write to the filesystem, execute shell commands for testing/verification, and interact with databases (SKILL.md, stack.yaml).
- Sanitization: The skill mitigates risk by enforcing clean architecture boundaries and the use of validation libraries (Zod, Pydantic) for processing data.
Audit Metadata