backend-dev
Pass
Audited by Gen Agent Trust Hub on Jul 31, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill follows security best practices for server-side development, instructing the agent to explicitly handle authorization, data invariants, and failure modes. No code, scripts, or external dependencies are included that would pose a security risk.
- [PROMPT_INJECTION]: The skill establishes an indirect prompt injection surface by defining workflows that ingest untrusted user data such as bug reports and feature requirements.
- Ingestion points: Defined in SKILL.md under 'Inputs to gather' and in examples/README.md within the 'Input' sections.
- Boundary markers: The skill does not define explicit delimiters or instructions to ignore embedded commands within the processed requirements.
- Capability inventory: The agent is granted capabilities to implement and refactor backend logic, modify data access layers, and generate tests (SKILL.md).
- Sanitization: No specific input validation or sanitization rules for the requirement text are provided.
Audit Metadata