T Backend Developer Guide
Welcome to T Backend! Your autonomous multi-agent AI platform is now ready for development. This guide will help you get started with the powerful features available.
🎯 What You Can Do
T Backend combines a conversational AI assistant with autonomous agents that can:
Research complex topics and provide detailed analysis
Monitor blockchain/crypto markets automatically
Process long-running tasks in the background
Manage knowledge graphs with temporal memory
Generate and edit images from text descriptions
Handle file uploads and create searchable document collections
📜 API Documentation
For complete details on all API endpoints, request formats, and responses, refer to our Swagger API Docs. This comprehensive documentation will guide you throu[...]
🔑 Authentication Setup
All API requests require an api-key header:
curl -H "api-key: YOUR_API_KEY_HERE" \
https://api-v2.fluxpointstudios.com/healthContact the Flux Point Team via email: [email protected] OR Discord: https://discord.gg/MfYUMnfrJM to get your API key if you need one.
Architecture & Execution Model
T Backend runs a privacy‑first, local‑execution model with optional overflow to cloud when needed. The assistant dynamically selects tools (web search, document search, image analysis) and compose[...]
Hybrid Memory (Graphs + Embeddings)
T uses a hybrid memory approach to improve context and recall:
Knowledge Graphs: Entities and relationships (with temporal context) captured from interactions; queryable via graph endpoints.
Semantic Embeddings: Conversation text and document snippets indexed for fast semantic recall (Document Q&A and
/chat).
This gives you: precise entity reasoning (graph) plus robust fuzzy matching and passage retrieval (embeddings).
Payments & Access Control (x402)
T Backend supports x402 “payment required” access control to gate API routes behind on‑chain proofs:
Pay‑to‑use flows for protected endpoints (e.g., premium analysis, data products).
Clean handoff from “402 Payment Required” to “200 OK” once a valid payment proof is submitted.
Designed to be composable with your app and simple to audit.
🌟 Core Features
1. Chat with the AI Assistant
The main chat endpoint provides conversational AI with advanced capabilities:
import requests
response = requests.post(
"https://api-v2.fluxpointstudios.com/chat",
headers={"api-key": "YOUR_API_KEY"},
json={
"message": "Research the latest developments in blockchain technology",
"session_id": "my-session-123" # Optional: maintains conversation context
}
)
print(response.json()["reply"])Features:
Remembers conversation context within sessions
Can process images (send
image_dataas base64)Uses external tools (web search, data analysis) automatically
Supports complex research queries
2. Background Tasks for Heavy Work
For complex analysis that might take several minutes:
# Start a background task
bg_task = requests.post(
"https://api-v2.fluxpointstudios.com/background/create",
headers={"api-key": "YOUR_API_KEY"},
json={
"model": "advanced", # Uses our most powerful AI model
"input": "Provide a comprehensive analysis of DeFi trends in 2025",
"analysis_depth": "high" # For complex analytical tasks
}
)
task_id = bg_task.json()["id"]
# Poll for results
import time
while True:
status = requests.get(
f"https://api-v2.fluxpointstudios.com/background/{task_id}",
headers={"api-key": "YOUR_API_KEY"}
).json()
if status["status"] == "completed":
print(status["output_text"])
break
elif status["status"] == "failed":
print("Task failed:", status.get("error"))
break
time.sleep(10) # Check every 10 seconds3. Autonomous Agent System
The platform includes specialized autonomous agents that work behind the scenes:
Available Agents:
Research Agent: Web research and fact-finding
Blockchain Agent: Crypto market monitoring
Consolidation Agent: Knowledge graph optimization
Maintenance Agent: System health monitoring
These agents automatically enhance your chat interactions and can be triggered through background tasks for specific analysis types.
General vs. Custom Endpoints
General endpoints (recommended starting points)
/chat: Conversational assistant (can ground answers in your documents/graphs)./background: Long‑running tasks with polling/streaming./files: File upload and vector store management./images: Image generation/editing; image analysis workflows./graph: Knowledge graph CRUD and querying.
Custom endpoints
For turnkey workflows (e.g., single‑purpose extract/verify pipelines), we can expose dedicated endpoints that compose multiple tools. These remain private to your integration.
4. File Upload & Search
Create searchable document collections:
# 1. Create a vector store for your documents
store = requests.post(
"https://api-v2.fluxpointstudios.com/files/vector-stores",
headers={"api-key": "YOUR_API_KEY"},
json={
"name": "my-documents",
"description": "My company's documentation"
}
).json()
store_id = store["id"]
# 2. Upload a file
file_response = requests.post(
"https://api-v2.fluxpointstudios.com/files/upload-url",
headers={"api-key": "YOUR_API_KEY"},
params={"url": "https://example.com/whitepaper.pdf"}
).json()
# 3. Add file to vector store
requests.post(
f"https://api-v2.fluxpointstudios.com/files/vector-stores/{store_id}/files",
headers={"api-key": "YOUR_API_KEY"},
json={"file_id": file_response["file_id"]}
)
# 4. Now you can search across your documents in chat!
search_response = requests.post(
"https://api-v2.fluxpointstudios.com/chat",
headers={"api-key": "YOUR_API_KEY"},
json={"message": "Search my documents for information about tokenomics"}
)Document Knowledge (Vector Stores)
Attach your own domain documents to T and query them:
Create private vector stores and upload documents.
Ask
/chattargeted questions; answers can be grounded in your sources.Stores are isolated per client; you decide which stores
/chatcan use on each request.
5. Image Generation & Editing
Generate and edit images from text:
# Generate a new image
image_response = requests.post(
"https://api-v2.fluxpointstudios.com/images/generate",
headers={"api-key": "YOUR_API_KEY"},
json={
"prompt": "A futuristic AI agent working with blockchain data",
"size": "1024x1024",
"quality": "high"
}
).json()
if image_response["success"]:
# Images are returned as base64 data
base64_images = image_response["images"]
print(f"Generated {len(base64_images)} images!")Image Analysis & OCR
T can analyze images and (optionally) run OCR for scanned documents to extract text fields. This enables:
License/document verification workflows.
OCR‑backed document extraction for scanned PDFs or images.
6. Knowledge Graph Access
Explore and query the temporal knowledge graph:
# List accessible graphs
graphs = requests.get(
"https://api-v2.fluxpointstudios.com/graph/accessible",
headers={"api-key": "YOUR_API_KEY"}
).json()
# Query a specific graph
if graphs["graphs"]:
graph_id = graphs["graphs"][0]["id"]
results = requests.get(
f"https://api-v2.fluxpointstudios.com/graph/{graph_id}/query?query=blockchain",
headers={"api-key": "YOUR_API_KEY"}
).json()🔧 Common Use Cases
Research & Analysis
# Deep research on a complex topic
research_task = requests.post(
"https://api-v2.fluxpointstudios.com/background/create",
headers={"api-key": "YOUR_API_KEY"},
json={
"input": "Research and analyze the impact of AI agents on financial markets",
"analysis_depth": "high"
}
)Document Q&A System
# Upload company docs to vector store, then ask questions
response = requests.post(
"https://api-v2.fluxpointstudios.com/chat",
headers={"api-key": "YOUR_API_KEY"},
json={"message": "Based on our uploaded documentation, what is our API rate limiting policy?"}
)Market Monitoring
# Ask about crypto trends (uses blockchain agent automatically)
market_response = requests.post(
"https://api-v2.fluxpointstudios.com/chat",
headers={"api-key": "YOUR_API_KEY"},
json={"message": "What are the current market conditions for Cardano and recent developments?"}
)Multi-turn Image Editing
# Generate an image, then refine it
initial_image = requests.post(
"https://api-v2.fluxpointstudios.com/images/generate",
headers={"api-key": "YOUR_API_KEY"},
json={"prompt": "A modern office workspace"}
).json()
# Edit the generated image
if initial_image["success"]:
edited_image = requests.post(
"https://api-v2.fluxpointstudios.com/images/edit",
headers={"api-key": "YOUR_API_KEY"},
json={
"prompt": "Add some plants and natural lighting",
"input_images": initial_image["images"]
}
)📊 Monitoring Your Usage
# Check background task history
tasks = requests.get(
"https://api-v2.fluxpointstudios.com/background/",
headers={"api-key": "YOUR_API_KEY"}
).json()
print(f"You have {len(tasks['responses'])} background tasks")
# View your vector stores
stores = requests.get(
"https://api-v2.fluxpointstudios.com/files/vector-stores",
headers={"api-key": "YOUR_API_KEY"}
).json()
for store in stores:
print(f"Store: {store['name']} (ID: {store['id']})")
# Check system health
health = requests.get(
"https://api-v2.fluxpointstudios.com/health",
headers={"api-key": "YOUR_API_KEY"}
).json()🛠️ Advanced Features
Background Task Types
You can create specialized background tasks for different agent types:
# Research-focused task
research_task = requests.post(
"https://api-v2.fluxpointstudios.com/background/create",
headers={"api-key": "YOUR_API_KEY"},
json={
"input": "research latest blockchain trends",
"task_type": "research"
}
)
# Consolidation task for knowledge optimization
consolidation_task = requests.post(
"https://api-v2.fluxpointstudios.com/background/create",
headers={"api-key": "YOUR_API_KEY"},
json={
"input": "consolidate recent knowledge and optimize storage",
"task_type": "consolidation"
}
)Streaming Results
For real-time updates on long-running tasks:
import requests
# Create a background task
task = requests.post(
"https://api-v2.fluxpointstudios.com/background/create",
headers={"api-key": "YOUR_API_KEY"},
json={"input": "Analyze market trends", "stream": True}
).json()
# Stream the results
response = requests.get(
f"https://api-v2.fluxpointstudios.com/background/{task['id']}/stream",
headers={"api-key": "YOUR_API_KEY"},
stream=True
)
for line in response.iter_lines():
if line:
print(line.decode('utf-8'))🚨 Important Notes
Rate Limits: Be mindful of API usage, especially with background tasks
Session IDs: Use consistent session IDs to maintain conversation context
Background Tasks: Use for analysis that takes more than 30 seconds
File Limits: Maximum 25MB per file upload
Model Choice: Use "advanced" model for complex reasoning tasks
API Keys: Keep your API keys secure and never commit them to version control
🆘 Troubleshooting
Common Issues:
401 Unauthorized: Check your API key is correct and active
403 Forbidden: You may not have access to that graph/resource
408 Timeout: Use background tasks for long-running operations
413 File Too Large: Files must be under 25MB
422 Unprocessable Entity: Check your request format matches the API specification
Getting Help:
Check the interactive API docs at
/docsendpointReview system health at
GET /healthContact the Flux Point Team via email: [email protected] OR Discord: https://discord.gg/MfYUMnfrJM for API key issues or permission problems
Debug Tips:
# Always check response status
response = requests.post(url, headers=headers, json=data)
if response.status_code != 200:
print(f"Error {response.status_code}: {response.text}")
else:
result = response.json()Data Isolation & Behavior Notes
Key‑scoped by design: everything stored (graphs, memories, vector knowledge) is isolated to your API key. No cross‑tenant leakage.
Duplicate & caching: by default, re‑sending the same document reprocesses it. Deterministic caching/dedupe can be enabled per workflow on request (e.g., cache‑by‑document ID with optional [...]
📚 Additional Resources
API Documentation: Visit
/docsfor interactive API explorer
🎉 You're Ready!
Your T Backend autonomous agent platform is ready for:
Intelligent conversations with memory
Deep research and analysis
Automated monitoring and insights
Document processing and search
Image generation and editing
Knowledge graph exploration
Start with simple chat requests and gradually explore the advanced features. The autonomous agents work behind the scenes to enhance your interactions automatically!
Quick Start Checklist:
Last updated