qa-workflow
Pass
Audited by Gen Agent Trust Hub on Mar 24, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The workflow involves running
npm test:e2efor regression testing and standardgitcommands for committing changes and merging to the master branch. - [INDIRECT_PROMPT_INJECTION]: The skill ingests data from
prd.jsonto track task status and blockers. This represents a potential indirect injection surface if the file contains untrusted content, though it is used here for routine state management. - Ingestion points:
prd.jsonreferenced in the status and blocking sections. - Boundary markers: No specific delimiters or safety warnings are provided for the data read from
prd.json. - Capability inventory: The skill possesses the ability to execute shell commands (
npm), perform file system operations (updating status files), and interact with remote repositories (git push/merge). - Sanitization: No explicit sanitization or validation of the input data from
prd.jsonis described in the workflow instructions.
Audit Metadata