software-development
Pass
Audited by Gen Agent Trust Hub on May 20, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill executes standard development tools including
gitfor version control and project-specific build scripts (compile, lint, test) as part of the commit verification process. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input in the form of software requirements and existing project code. This is an inherent surface for a software development tool.
- Ingestion points: Requirements provided during Phase 1 (Planning) and source code read during Phase 2 (Development).
- Boundary markers: No explicit delimiters are used to separate user instructions from system prompts.
- Capability inventory: Access to
gitcommands and arbitrary project build scripts (SKILL.md). - Sanitization: No explicit sanitization or validation of input data is performed.
Audit Metadata