API Documentation
Complete reference for the Veriprompt AI Gateway & Agentic Platform API
Categories
Quick Info
https://your-domain.comapplication/jsonPrompt ID System
{{variable_name}}- Centralized prompt management
- Version control for prompts
- Dynamic variable injection
- Team collaboration
Getting Started
1. Authentication
Get your API key from the dashboard and include it in all requests:
Authorization: Bearer YOUR_API_KEY2. Quick Test
Test the gateway with a simple prompt execution:
curl -X POST https://your-domain.com/api/gateway/execute -H "Authorization: Bearer YOUR_API_KEY" -H "Content-Type: application/json" -d '{"prompt": "Hello AI!", "quality": "fast"}'💡 Pro Tips
- Use prompt_id instead of raw prompts for better management
- Set quality parameter to balance cost vs performance
- Monitor usage via
/api/v1/analytics/overview - Use streaming for real-time chat applications
All Endpoints
39 endpoints found
/api/gateway/executeExecute Prompt
Execute a prompt through the AI Gateway with intelligent provider routing
/api/gateway/healthGateway Health
Check the health status of all AI providers and models
/api/gateway/injection-detectionInjection Detection
Detect potential prompt injection attacks
/api/v1/analytics/overviewAnalytics Overview
Get comprehensive analytics data for the organization
/api/v1/analytics/metricsAnalytics Metrics & Time Series
Retrieve success rate, cost, latency, and token time-series metrics for dashboards
/api/v1/analytics/exportAnalytics Export
Download execution, provider, or routing analytics as CSV or JSON
/api/v1/providers/credentials/analyticsBYOK Credentials Analytics
Get per-credential usage analytics, cost tracking, and performance metrics for Bring Your Own Key credentials
/api/v1/usageUsage Statistics
Get detailed usage statistics and costs
/api/v1/monitoring/dashboardMonitoring Dashboard
Real-time monitoring data for system health
/api/v1/monitoring/alertsMonitoring Alerts
Retrieve cost, performance, and usage alerts for the organization
/api/v1/monitoring/alertsAcknowledge or Dismiss Alert
Acknowledge or dismiss an alert from the monitoring system
/api/chat/streamStream Chat
Start a streaming chat conversation with AI models
/api/v1/chat/conversationsList Conversations
Get all chat conversations for the user
/api/v1/chat/conversations/[conversationId]Get Conversation
Get a specific conversation and its messages
/api/v1/promptsList Prompts
Get all prompts in the repository with filtering and pagination
/api/v1/prompts/fetchFetch Prompt Version
Retrieve the authorized prompt content for external execution without routing through the gateway.
/api/v1/prompts/exportExport Prompts (Offline Bundle)
Generate an encrypted offline bundle containing approved prompt content, attachments, and metadata for disaster recovery.
/api/v1/prompts/[promptId]/promotePromote Prompt Version to PROD
Promote a stored prompt version to the PROD channel.
/api/admin/offline-export-keysList Offline Export Keys
View active/previous offline export keys, download history, and recent offline export jobs.
/api/admin/offline-export-keysGenerate Offline Export Key
Rotate or create a new offline export key. Returns the plaintext key and one-time download token.
/api/admin/offline-export-keys/downloadDownload Offline Export Key
Redeem a one-time download token to fetch the encrypted offline export key.
/api/v1/promptsCreate Prompt Template
Create a new prompt template in the repository with variable placeholders
/api/v1/prompts/[promptId]/executeExecute Prompt by ID
Execute a specific prompt from the repository with variable injection
/api/v1/synthetic/providersSynthetic Providers
Get synthetic testing results for all providers
/api/v1/synthetic/analyticsSynthetic Analytics
Get analytics from synthetic prompt testing
/api/v1/workflows/statusIntegrated Workflow Status
Aggregate status across Git repositories, synthetic prompts, and analytics
/api/admin/provider-configsList AI Providers
Get all configured AI providers and their models
/api/admin/provider-configs/importImport Providers
Import AI providers from Excel/CSV file
/api/admin/usersList Users
Get all users in the system
/api/admin/analyticsAdmin Analytics
Get system-wide analytics and usage statistics
/api/user/profileUser Profile
Get current user profile information
/api/user/profileUpdate Profile
Update user profile information
/api/company/infoCompany Info
Get current company information and settings
/api/company/updateUpdate Company
Update company information and settings
/api/teamsList Teams
List teams the authenticated user belongs to with project counts
/api/teamsCreate Team
Create a new collaboration team for the organization
/api/teams/{teamId}Team Detail
Retrieve detailed information about a specific team including linked projects and members
/api/v1/routing-advisoryRouting Advisory
Get intelligent routing recommendations
/api/prompt-optimization/optimizeOptimize Prompt
Rewrite a prompt for clarity and structure via the Guru prompt-optimizer service pool. This is the single endpoint behind the Guru optimizer page, the chat terminal composer's optimize wand, and the Studio prompt editor's "Optimize with Guru" toolbar action — all three call it with identical behavior (same system prompt, sanitization policy, and usage/oversight logging). Anonymous callers get a rate-limited guest quota; authenticated calls are unlimited and logged against the company.