Platform Documentation
Everything you need to build, deploy, govern, and monitor AI agents on the TSI Agentic Hub.
Quick Start
# Clone and install
git clone <repository-url> agentic-platform
cd agentic-platform && pnpm install
# Start infrastructure + seed database
docker compose up -d
cd services/api-gateway && pnpm seed && cd ../..
# Start all applications
pnpm dev
Reference Documentation
Architecture
Full 10-layer architecture with data flows and technology choices
Protocols (A2A, AG-UI, A2UI, MCP)
How the 4 interoperability protocols work together with governance
API Reference
Complete REST API reference for all endpoints
SDK Guide (Python & TypeScript)
Python and TypeScript SDK reference with examples
Data Models & Types
All TypeScript types, database schemas, and enums
LLM Provider (AIFS)
T-Systems AIFS configuration, models, and SDK usage
A2A with Governance Extension
Turn-based execution protocol with governance layer
Configuration Reference
Environment variables, Docker, Kubernetes, and feature flags
API Explorer
Interactive API explorer — browse endpoints, test requests
Builder Guides
Registering Agents
Creating, configuring, and deploying agents in Agent Studio
Platform-Managed Agents
Build agents using the platform's built-in LLM runtime
External Agents (Pro-Code)
Connect LangGraph, CrewAI, PydanticAI, AutoGen, and custom agents
A2A Protocol & Client SDK
Implement the A2A protocol with governance extension
Tool Integration
Register MCP tools, configure tool sets, and grant tool access
Skills & Code Execution
Create skill scripts with 7-layer sandbox security
Skill Bundle Format
SKILL.md frontmatter, scripts/references/assets layout, agentskills.io compatibility
Knowledge Bases & RAG
Create knowledge bases, upload documents, configure retrieval
Memory Configuration
4 memory types, namespaces, TTL, and privacy controls
Access Control (Builder)
Configure all 4 access control dimensions for your agents
Eval Suite & Playground
Test agents with golden test sets and the prompt playground
Deployment & Monitoring
Deploy pipeline, canary releases, monitoring, and traces
Agent Design Patterns
When to use which runtime, single vs multi-agent patterns
Framework Integration
LangGraph, CrewAI, PydanticAI, AutoGen, n8n guidance
OBO Token Exchange
RFC 8693 delegated authentication for agent-to-service calls
Admin Guides
Identity & Users
IdP configuration, user provisioning, and role management
Model Governance
Model access policies, data sovereignty, and cost caps
Access Control (4D Model)
4-dimension access control with Cedar policy examples
Security & Compliance
SPIFFE identity, Sigstore, Cedar policies, audit trail
Guardrails Configuration
Input/output sensitivity thresholds and persistence
DSAR & Data Privacy
Data subject access requests and GDPR compliance
FinOps & Cost Management
Cost attribution, budgets, model tiers, and chargeback
Execution Management
Stop agents, set limits, monitor running executions
Troubleshooting
Common issues, error messages, and solutions
Best Practices
Applications
Agent Portal
localhost:3000 · Business Users
Chat with agents, manage approvals, view dashboards
Agent Studio
localhost:3001 · Builders / Engineers
Design, configure, test, and deploy agents
Admin Console
localhost:3002 · Platform Engineers
Manage tenants, policies, FinOps, and infrastructure
Docs Portal
localhost:3003 · All Users
Developer documentation, API reference, guides, and best practices