spec-kit-specify
Pass
Audited by Gen Agent Trust Hub on Apr 13, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local shell script at
scripts/create-new-feature.shto initialize feature branches and manage project metadata. - [PROMPT_INJECTION]: The skill processes untrusted user requirements and interpolates them into script arguments and generated specification files, presenting an indirect prompt injection surface.
- Ingestion points: Full user request captured as feature description in
SKILL.md(Workflow Step 01). - Boundary markers: No explicit delimiters are used to wrap or sanitize user input in the generated
spec.mdor script calls. - Capability inventory: Shell command execution via local scripts and file system writes to the repository.
- Sanitization: No validation or escaping of the user-provided feature description is performed.
- [EXTERNAL_DOWNLOADS]: References a specification template from an official repository on GitHub.
Audit Metadata