backend-development-feature-development
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user data which is interpolated into prompts for specialized subagents, potentially allowing malicious instructions in the input to influence downstream actions.
- Ingestion points: User-provided feature descriptions (passed through the
$ARGUMENTSvariable) are interpolated into the instructions for every phase of the workflow inSKILL.md. - Boundary markers: The prompts do not use explicit delimiters or "ignore instructions" warnings to separate the trusted subagent instructions from the untrusted user input.
- Capability inventory: The subagents invoked (such as
backend-architect,deployment-engineer, anddata-engineerinSKILL.md) possess high-privilege capabilities including backend code generation, database migration creation, and CI/CD pipeline configuration. - Sanitization: There is no evidence of validation or sanitization of the user input before it is passed to the subagents.
Audit Metadata