trinity

[!NOTE] 📦 Just released: v0.9.5 (September 17, 2026). Workspace is where the work happens now:

• chats per topic
• a conversation rail
• live execution cards
• agents that ask you questions
• a canvas for every agent
• real-time voice mode

Also new:

• prebuilt images and a guided DigitalOcean install
• first-run setup in the browser
• subscription usage and headroom you can see
• opt-in instance telemetry
• credentials encrypted at rest

What's new overview → · Full release notes → · Previous: v0.9.0 release review video →

---

The Sovereign AI Agents Platform

Trinity is a sovereign AI agents platform — build and run proprietary systems, from fully (or semi-) autonomous businesses to self-improving cognitive systems, all governed, auditable, and on infrastructure you control.

Claude Code writes the agent. Trinity runs it in production.

Each agent runs in its own isolated Docker container with real-time observability, fleet-wide scheduling, agent-to-agent delegation, and a tamper-evident audit trail. Self-host it, or run it on any cloud you control.

Open source · Apache 2.0 — free for any use, commercial included, and deploys anywhere you run it · Independently pentested — UnderDefense Grade A · We run Trinity in production ourselves — and so do our customers.

🤖 AI agent reading this repo? Start at AGENTS.md — a task router with exact commands, key facts, and verification steps. Detailed docs index: docs/user-docs/README.md.

[!IMPORTANT] PostgreSQL is the recommended database for production. SQLite stays the zero-config default for local development and evaluation, but production instances should run on PostgreSQL — opt in with a single DATABASE_URL (setup guide, #300). Already on SQLite? Migrate with the Trinity Ops Agent's /migrate-to-postgres skill — a gated, validate-then-cutover flow that leaves your SQLite file untouched for instant rollback.

Why Trinity?

The problem: Everyone wants autonomous AI agents. But your options are terrible — SaaS platforms where data leaves your security perimeter, custom builds that take 6-12 months, or frameworks that don't handle governance and audit trails.

The solution: Trinity is autonomous agent orchestration infrastructure. Each agent runs in its own isolated Docker container. You get real-time observability, fleet-wide health monitoring, cron-based scheduling, agent-to-agent delegation, and cost tracking — all on your own hardware.

Option	Problem	Trinity
SaaS Platforms	Data leaves your perimeter, vendor lock-in	Your infrastructure, data never leaves
Build Custom	6-12 months, $500K+ engineering	Deploy in minutes
Frameworks	No observability, no fleet management	Real-time monitoring, scheduling, audit trails

See it move

A short teaser of the live interface — the full, zoomable detail is in the screenshots below.

Watch more:

• Trinity v0.9.0 Release Review — everything new in v0.9.0, reviewed live in the "How to Build a Second Brain for Your AI Agents" workshop
• Trinity Demo — full platform walkthrough
• Trinity Architecture Explained — animated explainer (4:44): one scheduled task followed end to end through the container, gate, delegation, approval, audit, and breaker machinery
• Trinity v0.8.0 Release Tour — agents that speak, the Brain Orb, and the Grid dashboard
• From Zero to Deployed AI Agent — your first Trinity agent in 30 minutes
• Plugins + Trinity: Build and Deploy Agents in Cursor — the plugin workflow end-to-end
• Loops Engineering with Trinity — bounded autonomous work loops in practice
• The Multi-Agent Platform I Run My Company On — live fleet tour: delegation, memory, observability
• Autonomous Cornelius on Trinity — a real second-brain agent running autonomously (Cornelius on GitHub)

More recordings — agent dev pipelines, GitHub-backed agents, an ops agent that deploys other agents — in the workshop archive. Live and free, every Thursday.

Up and running in minutes, not months

01 · Deploy an instance	Self-host with one command — your own machine, any server you control, or a DigitalOcean Droplet in about ten minutes.
02 · Connect Claude Code	Install the abilities plugins — scaffold, connect, and deploy agents over MCP.
03 · Run in production	Scheduled, multi-user, audited — inside your own perimeter.

---

Quick Start

Two phases: stand up an instance (Phase A), then build and deploy agents to it from Claude Code (Phase B).

Phase A — Stand up a Trinity instance

Prerequisites

• Docker and Docker Compose v2+
• Anthropic API key (for Claude-powered agents) OR Google API key (for Gemini-powered agents)

Recommended: let your agent install it (one shot)

Tell your Claude (or any capable coding agent) to install Trinity and follow the agent-facing runbook — it verifies Docker, installs unattended, confirms the stack is serving, surfaces the generated admin password, and hands you a next-steps card:

Install Trinity on my computer. Follow the runbook at https://raw.githubusercontent.com/abilityai/trinity/main/docs/AGENT_INSTALL_GUIDE.md

The runbook (docs/AGENT_INSTALL_GUIDE.md) is a deterministic verify → install → confirm → report-next-steps loop that any capable agent can execute and explain to you as it goes.

Install it yourself

# 1. Clone the repository
git clone https://github.com/abilityai/trinity.git
cd trinity

# 2. Configure environment
cp .env.example .env
# Edit .env and set ADMIN_PASSWORD (12+ characters) — the only required edit.
# start.sh generates SECRET_KEY, the encryption keys and the Redis
# passwords for you on first run.

# 3. Build the base agent image
./scripts/deploy/build-base-image.sh

# 4. Start all services
./scripts/deploy/start.sh

For a fully non-interactive bring-up, run ./scripts/deploy/start.sh --unattended (or set TRINITY_UNATTENDED=1) and skip the .env edit entirely — an admin password is generated and printed in the final summary.

Note: the repo's git submodules are private and optional — cloning (even with --recurse-submodules) needs no credentials; they're skipped automatically. See docs/ENTERPRISE.md.

First-time setup

1. Open http://localhost and log in with the username admin and the ADMIN_PASSWORD from .env.
2. First-run setup opens on the Dashboard. The one required step is Connect Claude: paste a Claude subscription token or an Anthropic API key, and Trinity checks it before saving.
3. Optional steps follow: GitHub, email-provider and Gemini keys, then your first agent. Anything you skip is under Settings → Integrations later.

Access

• Web UI: http://localhost
• API Docs: http://localhost:8000/docs
• MCP Server: http://localhost:8080/mcp

Don't want to run it on your laptop? Trinity is self-hosted, and the least work is a server of your own: the guided DigitalOcean install gets you a Droplet behind HTTPS in about ten minutes. Questions about a deployment? Talk to an engineer → — an engineer reads this, not a CRM. Reply in one business day, your time zone.

On DigitalOcean? The guided installer (scripts/deploy/trinity-do-create.sh) takes a few steps and about ten minutes. You install doctl, create a Claude subscription token, choose an admin password, and run one command. The result is Trinity on its own Droplet behind HTTPS. See Deploy on DigitalOcean.

Prebuilt images: ./scripts/deploy/start.sh --hosted pulls the published GHCR images instead of building from source. Pin TRINITY_IMAGE_TAG (for example v0.9.5) on any instance you keep.

Deploying to a remote server? /trinity:deploy-new-instance from the abilities marketplace provisions Trinity on any server you can SSH into — and scaffolds the Trinity Ops Agent to manage it (health, logs, updates, rollback, and SQLite→PostgreSQL migration).

Phase B — Build & deploy agents from Claude Code

Most of the Trinity workflow lives in the abilities plugin marketplace — Claude Code plugins covering the full agent lifecycle: scaffold, develop, deploy, iterate. You need the instance from Phase A (localhost or remote) and Claude Code.