ai-assisted-testing
Pass
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill includes several Python scripts (e.g.,
scripts/batch_convert_templates.py,scripts/convert_to_json.py) that usesubprocess.callto invoke internal helper scripts for file format conversion. These commands are implemented usingsys.executableand list-based arguments, which avoids shell-related injection risks. - [INDIRECT_PROMPT_INJECTION]: This skill is intended to process untrusted project data such as requirements and API documentation, presenting a surface for indirect prompt injection.
- Ingestion points: External data is ingested as described in
prompts/ai-assisted-testing.md(e.g., requirements, user stories, API docs). - Boundary markers:
SKILL.mdincludes instructions to classify and sanitize input, replacing sensitive data with placeholders. - Capability inventory: The skill is capable of local file system operations and internal command execution, but lacks network capabilities.
- Sanitization: The workflow requires human review of all AI-generated drafts and explicit acceptance before they are used as execution assets.
Audit Metadata