Salesforce Sandbox Data Seeder
Fills a Salesforce sandbox with realistic, relationship-aware data that satisfies the org's own validation rules.
- Status
- Running
- Timeline
- Aug 2025 to 19 Aug 2026
- Built with
- Node.js, Salesforce APIs, Anthropic / OpenAI-compatible / Ollama, Docker
- Links
- Source on GitHub
Empty sandboxes make for poor testing, and generic fake data fails the moment it meets a real org’s picklist dependencies and validation rules.
What it does
Point it at an org, pick the objects, and it generates records that respect the org’s actual metadata: dependent picklists, field lengths and precision, restricted values, and lookups loaded in dependency order with real record ids.
It reads active validation rules, has an AI model translate them into field constraints, and generates records that pass. Failed records are repaired by error code and retried.
Run it your way
A guided seven-step web wizard or a CLI. Bring your own model (Anthropic, any OpenAI-compatible endpoint, or local Ollama), or run without AI on pattern-based generation. Ships as Docker images.