qa-test-planner
Pass
Audited by Gen Agent Trust Hub on Mar 18, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: No evidence of instructions designed to bypass safety filters, override system prompts, or extract internal instructions was found. The instructional language is focused on standard QA workflows.
- [COMMAND_EXECUTION]: The skill includes two local bash scripts (
scripts/generate_test_cases.shandscripts/create_bug_report.sh) used for interactive documentation generation. These scripts use standard shell commands to capture user input and write results to Markdown files locally. - [DATA_EXFILTRATION]: No suspicious network operations or unauthorized access to sensitive files (such as SSH keys or cloud credentials) were identified. The data handling is confined to generating documentation based on user-provided feature descriptions.
- [REMOTE_CODE_EXECUTION]: There are no patterns involving the download and execution of remote scripts (e.g.,
curl | bash) or any attempts to run unverified external code. - [EXTERNAL_DOWNLOADS]: The skill references Figma MCP for design validation. This is a recognized service integration used for legitimate design-to-code verification and does not involve downloading untrusted or malicious payloads.
- [INDIRECT_PROMPT_INJECTION]: While the skill ingests external data (user requirements and Figma design specs) to generate documentation, it lacks high-risk capabilities like dynamic code execution or automated system modifications that would make this a critical vulnerability surface. It functions as a text-generation assistant.
Audit Metadata