Skip to main content
Trinity

Guides

Step-by-step tutorials for common tasks and advanced configurations. Each guide walks you through a complete workflow from start to finish.

Creating Agents

Create agents from templates, GitHub repos, or from scratch

Managing Agents

Start, stop, rename, delete agents and monitor health

Agent Chat

Chat interface, voice input, streaming, and conversation history

Agent Terminal

Browser-based xterm.js terminal with Claude, Gemini, and Bash modes, plus SSH access

Agent Files

File browser, virtual filesystem, and shared folders

Agent Logs

Log viewing, telemetry, and Vector aggregation

Agent Configuration

Autonomy levels, read-only mode, resources, and timeout settings

Credential Management

Securely inject and manage agent credentials

OAuth Credentials

OAuth2 flows for Google, Slack, GitHub, and Notion

Subscription Credentials

Shared Claude subscriptions, auto-assign, and auto-switch

MCP Integration

Connect Claude Code and other MCP clients to Trinity

Scheduling

Cron schedules, execution queue, and misfire handling

Skills and Playbooks

Skills library, assignment, and chat autocomplete

Approvals

Human-in-the-loop approval gates for agent actions

GitHub Sync

Keep agents in sync with GitHub repositories

Slack Integration

Connect agents to Slack with DMs, mentions, and multi-agent routing

Nevermined Payments

x402 payment protocol for agent monetization

Agent Sharing

Share agents with other users and set access levels

Public Links

Public chat URLs, email verification, and session memory

Tags and Organization

Tags, filtering, and system views

Mobile Admin

Mobile PWA for agent management at /m

Advanced Features

Voice chat, image generation, process engine, and dynamic dashboards

Voice Chat

Real-time voice via Gemini Live API

Image Generation

Gemini two-step image pipeline for agents

Agent Avatars

AI-generated avatars with emotion variants

Process Engine

BPMN workflows, step types, and process analytics

Dynamic Dashboards

Custom agent dashboards via YAML configuration

Agent Compatibility Checklist

Dockerfile with SSH server (port 22) and agent-server.py
.env file for credential injection
.mcp.json.template with ${VAR} placeholders
Health check endpoint responding on the internal server