CrewCheck + LlamaIndex Integration
Integrate CrewCheck with LlamaIndex for DPDP-compliant AI governance. Step-by-step setup guide for ai framework integration with PII redaction and audit trails.
Overview
Integrate CrewCheck with LlamaIndex to add DPDP-compliant AI governance to your ai framework workflows. LlamaIndex RAG pipelines can route through CrewCheck by configuring the LLM provider's base URL. Your index, retriever, and query engine code remains unchanged.
Quick setup
``` # In your LlamaIndex config llm = OpenAI(api_base='https://your-org.crewcheck.in/v1/proxy/openai') ```
Once configured, all LlamaIndex traffic passes through CrewCheck's governance layer. Indian PII (Aadhaar, PAN, UPI, mobile numbers) is automatically detected and redacted. Policy packs are evaluated on every request. Every interaction is logged to the tamper-evident audit trail.
What you get
• Automatic Indian PII detection and redaction for all LlamaIndex interactions
• DPDP Act 2023 compliance controls applied at the request boundary
• Real-time cost tracking in ₹ for every LlamaIndex API call
• Tamper-evident audit trail for regulatory reporting
• Shadow Mode to test governance controls before enforcement
• Dashboard visibility into LlamaIndex-specific metrics and compliance scores
Next steps
After setting up the LlamaIndex integration, configure your policy packs in the CrewCheck dashboard. Start with Shadow Mode to observe detection rates and policy impacts, then promote to enforcement when you're confident in the configuration.
For production deployments, enable circuit breakers to automatically halt LlamaIndex traffic if PII detection rates or costs exceed your configured thresholds.