https://api.dashboard.valuez.ai
All requests require the x-api-key header containing the API key provided by your workspace admin. See API Authentication.
Agent & KB Lifecycle
1. Creating & Managing Agents (/api/v1/external/agent/*)
1.1 List Workspace Agents
Fetch all existing agents with pagination and sorting.- Endpoint:
POST /api/v1/external/agent/agent-list
1.2 Create or Update an Agent
Creates a new AI agent or updates an existing one.- Endpoint:
POST /api/v1/external/agent/add-update-agent
Field Explanations:
1.3 Fetch a Single Agent
- Endpoint:
GET /api/v1/external/agent/get-agent?AgentKeyID={AgentKeyID} - Returns full agent metadata, system prompt, greetings, and associated KB ID.
2. Voice & Model Configuration (/api/v1/external/agent-config/*)
Once an agent is created, configure its runtime engines: Speech-to-Text (STT), Large Language Model (LLM), and Text-to-Speech (TTS).
2.1 Discover Voice & LLM Models
Before setting configuration, query the available models for your tenant:GET /api/v1/external/agent-config/llm-provider-list- Foundation LLM providers (Sarvam, OpenAI, etc.).GET /api/v1/external/agent-config/voice-provider-list- TTS engines.GET /api/v1/external/agent-config/voice-model-list?voiceProviderID=12- Specific voices available for a provider.GET /api/v1/external/agent-config/transcribe-model-list?transcribeProviderID=8- STT models.GET /api/v1/external/agent-config/demo-voice?voiceID=1&voiceProviderID=12&speed=1.00- Audio preview stream.
2.2 Save Runtime Configuration
- Endpoint:
POST /api/v1/external/agent-config/insert-agent-config
3. Knowledge Base Setup (/api/v1/external/kb/*)
A Knowledge Base (KB) powers the Retrieval-Augmented Generation (RAG) system. Every agent links to a KB ID.
3.1 Create or Update a Knowledge Base
- Endpoint:
POST /api/v1/external/kb/add-update-kb
- When creating, pass
"KBKeyID": null. The response returns the newly createdKBKeyIDand internalKBID. ConversationalModeIDs: String of comma-separated modes (1= Voice,2= Chat).
3.2 Clear Vector Embeddings
- Endpoint:
POST /api/v1/external/kb/clear-qdrant - Destructively wipes vector embeddings in Qdrant for a collection (e.g.,
kb_embeddings_{OrgID}_{KBID}). Used during full re-indexing.
4. Ingesting Knowledge Data
A Knowledge Base can ingest four types of data:4.1 Documents (PDF, Excel, Word) - /api/v1/external/kb-doc/*
Upload company policies, product sheets, or manuals. The backend chunks the document, generates embeddings, and indexes them in Qdrant.
- Endpoint:
POST /api/v1/external/kb-doc/add-update - Content-Type:
multipart/form-data
4.2 Frequently Asked Questions (FAQs) - /api/v1/external/kb-faq/*
Direct Q&A pairs provide deterministic, high-confidence answers for common caller inquiries.
- Endpoint:
POST /api/v1/external/kb-faq/add-update-faq
KBFAQID: 0 for new questions; pass existing integer ID to edit).
4.3 Business Profile & Hours - /api/v1/external/kb-business-info/*
Gives the agent foundational facts about your business, office addresses, and operating hours.
- Endpoint:
POST /api/v1/external/kb-business-info/add-update
4.4 Product & Service Catalog - /api/v1/external/kb-catalog/*
Allows agents to quote prices, check inventory availability, and explain service terms.
- Endpoint:
POST /api/v1/external/kb-catalog/add-update
5. Inspecting Conversation Logs (/api/v1/external/logs/*)
Track conversations handled by your agents:
- List Logs:
POST /api/v1/external/logs/conversational-log-list
- Get Transcript Detail:
GET /api/v1/external/logs/get-conversational-log?ConversationalLogKeyID={uuid}
Speech & Agent APIs
Test your agent with live chat queries.
Launch Outbound Campaigns
Deploy this agent in automated calling campaigns.
