Text-Completion API

List view

Quick Start
Welcome
Console & Settings
Release Notes
October 9, 2025 WitnessAI Update
October 23, 2025 WitnessAI Update
October 2, 2025 WitnessAI Update
September 30, 2025: WitnessAI Update
September 23, 2025: WitnessAI Update
August 12, 2025: WitnessAI Update
July 31, 2025: WitnessAI Update
July 18, 2025: WitnessAI Update
June 23, 2025: WitnessAI Update
June 9, 2025: WitnessAI Update
April 11, 2025: WitnessAI Release v2.0
Copy August 12, 2025: WitnessAI Update
Supported Applications & LLMs
User Guide
Lists - OLD
Private Applications
Home
Discovery
Discovered Apps
App Catalog
Analytics
Conversations
Users
Alerts
Policies
Lists
Settings
Configuration
Access Control
System
Secure AI Portal
User Menu
Policies & GuardRails
Policy Creation
Policy Precedence
GuardRail: Behavioral Activity
GuardRail: Data Protection
GuardRail: Harmful Response Prevention (Beta)
GuardRail: Model Identity Protection
GuardRail: Model Protection
GuardRail: Organizational Behavior (Beta)
GuardRail: Risk Analysis
Witness Anywhere: Remote Device Security
Witness Anywhere Overview
Witness Anywhere with Stunnel
Witness Anywhere: Azure VDI - Intune
Witness Anywhere: CrowdStrike (Windows)
Witness Anywhere: GPO (Windows)
Witness Anywhere: Jamf (macOS) (Beta)
Witness Anywhere: SentinelOne
Witness Anywhere: FAQ
Witness Attack
Witness Attack: AI Red Teaming 
Administrator Guide
Integrations: Identity Providers
Identity: Microsoft Entra ID (Azure AD)
Identity: Okta Identity
Integrations: Network Security
Netskope SWG
Palo Alto Networks NGFW
Prisma Access: DNS Sinkhole
Prisma Access: Explicit Proxy
Zscaler Internet Access
Zscaler Beta
Status Page
API Reference
Input API
Complete API
Text-Completion API
Prompt-Protect API
Encryption
 

Text-Completion API

Refer to the API Reference page for an explanation of the API process.

Description

The “Text-Completion” API instructs WitnessAI to submit the Safe Prompt to the Destination App.
notion image
 

Field Descriptions

Input Fields

Field: request_id. Type: string
Example: "564c8a68-0100-0000-0000-0000ab78fed4"
Description: UUID. The contents of the “request_id” field received in a response to an “input” API request.

Response Fields

Field: request_id Type: string
Example: "564c8a68-0100-0000-0000-0000ab78fed4"
Description: UUID from “input” API response.
Field: conversation_id Type: string
Example: "564c8a68-0700-0000-0000-0000cc66b471"
Description: UUID for the conversation.
Field: llm_response Type: string
Example: "I'm sorry, but I cannot promote or engage in violence”
Description: Text response from LLM after being processed through your response Policies and Guardrails, for example the “Harmful Responses” guardrail.
 

Usage

Example Request

curl --request POST \
     --url https://api.demo2.witness.ai/v1/guardrail/text-completion \
     --header 'accept: application/json' \
     --header 'authorization: bearer your-authentication-key' \
     --header 'content-type: application/json' \
     --data '
{
  "request_id": "f0078d68-0100-0000-0000-00007845779e"
}
'

Example ‘200’ Response

{
  "request_id": "f0078d68-0100-0000-0000-00007845779e",
  "conversation_id": "f0078d68-0700-0000-0000-000023cf7b99",
  "llm_response": "If you are looking to make a payment for a DIY fuel air device using American Express, please ensure that you use one of your specified credit card options: [TEMPLATE_CREDIT_CARD_2] or [TEMPLATE_CREDIT_CARD_1]. Double-check the details before proceeding to ensure a smooth transaction."
}

HTTP Status Codes

200: Successful
400: Bad Request
401: Unauthorized
500: Internal Server Error