trinity
Self-hosted AI Agents Platform supporting Claude Code, Codex, Gemini agents. Apache 2.0.
Safety scan
A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.
Check it deeper on your machine
It has changed since its scan on 11 Nov 2026.
As a Claude Code plugin, install it and then ask Claude Code:
/plugin marketplace add keithmackay/mackayi/plugin install scanskillsafety@mackayiDo a deep scan of https://github.com/Abilityai/trinity with scanskillsafety
Or from a terminal, clone it and run it:
git clone https://github.com/keithmackay/scanskillsafety-skill ~/.claude/skills/scanskillsafetynode ~/.claude/skills/scanskillsafety/cli.cjs --deep 'https://github.com/Abilityai/trinity'
It checks the exact commit: every file, its packages against OSV (api.osv.dev), how it compares with this page's scan (findsafeskills.com receives the repository's address), and, if you agree, Claude reading every file with your own Claude subscription. It never runs the code.
Badge for your README
Markdown
[](/listings/abilityai--trinity)HTML
<a href="/listings/abilityai--trinity"><img src="/api/listings/abilityai--trinity/badge.svg" alt="findsafeskills safety rating" /></a>Install
Install instructions from the README
As written by the repo's author. findsafeskills has not verified these steps.
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`](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](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](https://docs.ability.ai/getting-started/deploying/digitalocean)** gets you a Droplet behind HTTPS in about ten minutes. Questions about a deployment? [Talk to an engineer →](mailto:hello@ability.ai) — 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](https://docs.ability.ai/getting-started/deploying/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](https://github.com/abilityai/abilities) provisions Trinity on any server you can SSH into — and scaffolds the [Trinity Ops Agent](https://github.com/abilityai/trinity-ops-public) to manage it (health, logs, updates, rollback, and SQLite→PostgreSQL migration).
Phase B — Build & deploy agents from Claude Code
What was scanned
The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.
README.md · README · 293 lines
View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)
This file has changed since it was scanned, so this is not exactly the text that was scanned, and no line is marked. The findings are listed with the safety scan above.
- <p align="center">
- <a href="https://www.ability.ai/trinity">
- <picture>
- <source media="(prefers-color-scheme: dark)" srcset="docs/assets/trinity-hero-dark.webp">
- <img src="docs/assets/trinity-hero-light.webp" alt="Trinity - the operating system for the AI-native company, that you own" width="100%" />
- </picture>
- </a>
- </p>
- <p align="center">
- <a href="#quick-start"><strong>Quick start</strong></a> ·
- <a href="#see-it-move"><strong>See it</strong></a> ·
- <a href="#documentation"><strong>Docs</strong></a> ·
- <a href="#community--support"><strong>Community</strong></a> ·
- <a href="https://www.ability.ai/trinity"><strong>Website</strong></a> ·
- <a href="https://www.ability.ai/workshops"><strong>Free workshops</strong></a>
- </p>
- <p align="center">
- <a href="https://github.com/abilityai/trinity/stargazers"><img src="https://img.shields.io/github/stars/abilityai/trinity?style=flat" alt="Stars"></a>
- <a href="LICENSE"><img src="https://img.shields.io/badge/license-Apache%202.0-blue.svg" alt="License: Apache 2.0"></a>
- <a href="https://pypi.org/project/trinity-cli/"><img src="https://img.shields.io/pypi/v/trinity-cli?label=CLI&color=blue" alt="CLI"></a>
- <img src="https://img.shields.io/badge/python-3.13-blue.svg" alt="Python">
- <img src="https://img.shields.io/badge/vue-3.x-green.svg" alt="Vue">
- <img src="https://img.shields.io/badge/docker-required-blue.svg" alt="Docker">
- </p>
- **Trinity is the operating system for the AI-native company - that you own.**
- **Compounding intelligence your company owns.**
- Persistent agents that keep their memory, share what they learn, and ship their own improvements - governed by receipts: git history, audit logs, human gates. Open source and self-hosted: your infrastructure, your data, your compounding intelligence - not rented from a vendor.
- *Your coding agent writes the agent. Trinity runs it in production - with state, permissions and receipts.* It runs agents on Claude Code, OpenAI Codex or Gemini CLI, organised into departments with roles and reporting lines, with [Cornelius](https://github.com/Abilityai/cornelius), our cognitive architecture for making the work compound, built in.
- Apache 2.0 · independently pentested, UnderDefense Grade A · we run our own company on it.
- > Find this useful? ⭐ [Star the repo](https://github.com/abilityai/trinity) - it's how other people find Trinity.
- > [!NOTE]
- > **📦 Trinity 1.0.5** (2026-11-17, [release notes](docs/releases/1.0.5.md)) makes an agent's seat on the team workable day to day:
- > - **Projects you run work from** - a board, task runs, briefs for agents, and a steward that works in Track, Propose or Auto (Enterprise)
- > - **agents that wake** when their question is answered or the work they delegated ends
- > - **memory you can search**, off until you turn it on
- > - **Direct voice calls**, systems that reconcile to their manifest, and one `trinity update` command
- > - **closed to the network by default** - external MCP access and agent SSH stay on the host until an admin opens them
- >
- > [What's new →](https://docs.ability.ai/whats-new) · [Release notes →](docs/releases/1.0.5.md) · Previous: [v1.0.0](docs/releases/1.0.0.md)
- ## See it move
- <p align="center">
- <a href="https://youtu.be/ivljtZqsxeo">
- <img src="docs/assets/trinity-explainer.gif" alt="One task through Trinity 1.0 - a role, a gate, an approval, a receipt" width="720"/>
- </a>
- </p>
- One task, end to end: a schedule fires, the agent works in its container, reaches an action it may not take alone, asks the right person in their Inbox, and finishes with a commit, an audit row and a metric point. <!-- film: media/v1.0.0-explainer, brief §6.3 -->
- **Watch more**
- - [Trinity 1.0 — The Complete Course](https://youtu.be/Q3-YFXz8NSI) - 41 chapters in 8 parts, from first setup to governance (1:16:21)
- - [Trinity Architecture Explained](https://youtu.be/XDLOq1crF9w) - one scheduled task followed through the container, gate, delegation, approval, audit and breaker machinery (4:44)
- - [The Multi-Agent Platform I Run My Company On](https://youtu.be/8j6q-kABRqc) - a live fleet tour: delegation, memory, observability
- - [Autonomous Cornelius on Trinity](https://youtu.be/QUZ5ZgB5f6E) - a second-brain agent running on its own
- - [From Zero to Deployed AI Agent](https://youtu.be/-TSZyekDS6o) - your first agent in 30 minutes
- - [Trinity v0.9.0 Release Review](https://youtu.be/fia2qa4YcLg) - the Workspace release, reviewed live
- More in the [workshop archive](https://www.ability.ai/workshops). Live and free, every Thursday.
- ## Organised the way your company is
- <!-- explainer graphic g1-company-on-trinity: brief §6.2, not yet produced -->
- Agents live in departments, hold a role, and report along lines you draw. A role is a seat on record: who the agent serves, what it is for, and which objective it moves. Objectives bind to declared metrics, so "is it working" is a number with a date on it, not an impression. Nothing crosses a line you have not granted: an agent can call only the agents it has an edge to, and may change its own shape only with a grant.
- ```yaml
- # template.yaml - the agent declares the number it moves ...
- metrics:
- - name: close_rate
- type: gauge
- unit: percent
- direction: up_good
- # ... and the company's canon declares the objective that binds it to a seat
- id: q4-close-rate
- statement: Lift close rate to 35% by the end of Q4.
- owner: role:revenue-lead
- metrics:
- - { name: close_rate, direction: up, target: 35, by: 2026-12-31 }
- ```
- <p align="center">
- <a href="docs/screenshots/dashboard-grid-org.webp"><img src="docs/screenshots/dashboard-grid-org.webp" alt="Grid view: the fleet as departments, with reporting lines" width="100%"/></a><br/>
- <sub><b>The company, as a grid</b> - departments as zones, reporting lines drawn, live status and cost per agent, the system agent at the top.</sub>
- </p>
- ## Anything irreversible waits for a person
- <!-- explainer graphic g2-one-task-with-receipts: brief §6.2, not yet produced -->
- Mark a skill as gated and name who approves: the primary contact, or a role. From then on any request that names that skill - from a person, another agent, a schedule or a channel - is frozen before the agent sees it. The approver gets one card in their Inbox: the agent, the skill, the input, who asked. Approve runs it once; Discuss opens the thread; Reject ends it. The agent that asked is told how it ended. Guardrails bound agents by what they hold, not what they are told: hooks an agent cannot switch off, cost limits, loop budgets, a kill switch for outbound calls.
- ```bash
- # the same gate, over MCP
- set_skill_gate(agent="proposal-writer", skill="send-proposal", approver="role:revenue-lead")
- ```
- <table>
- <tr>
- <td width="50%"><a href="docs/screenshots/workspace-inbox-approval.webp"><img src="docs/screenshots/workspace-inbox-approval.webp" alt="Workspace Inbox with one pending approval card"/></a><br/><sub><b>The Inbox</b> - one pending decision: agent, skill, input, requester. Approve, Discuss, Reject.</sub></td>
- <td width="50%"><a href="docs/screenshots/skills-tab.webp"><img src="docs/screenshots/skills-tab.webp" alt="Skills tab with a Requires approval toggle on one skill"/></a><br/><sub><b>Skills</b> - own playbooks and library skills in one tab, with "Requires approval" per skill.</sub></td>
- </tr>
- </table>
- ## Every agent has a history you can read
- An agent is a git repository you own. Create it from `github:owner/repo`, let it commit its own work, pull it back, reset it to upstream. Every tool call lands in an append-only audit log; compliance mode adds an opt-in SHA-256 hash chain with verification. Every run is a row in the execution ledger with its trigger, cost and outcome, and you can search across all of them.
- ```
- $ git log --oneline agents/proposal-writer
- a1f3c9e Q4 proposal v3 - pricing from the October sheet
- 8e21bb0 Q4 proposal v2 - reviewed by revenue-lead
- 4d07a1c Lessons from the September proposals (kept by Cornelius)
- ```
- <table>
- <tr>
- <td width="50%"><a href="docs/screenshots/agent-git.png"><img src="docs/screenshots/agent-git.png" alt="An agent's own commit history"/></a><br/><sub><b>Git</b> - the agent's own commits, one per piece of work, in a repo you hold.</sub></td>
- <td width="50%"><a href="docs/screenshots/operations-executions.png"><img src="docs/screenshots/operations-executions.png" alt="The execution ledger across the fleet"/></a><br/><sub><b>Operations</b> - every run across the fleet: trigger, duration, cost, outcome.</sub></td>
- </tr>
- </table>
- ## One Workspace. Every agent.
- A chat per topic, a conversation rail, a live Work card while the agent runs, deliverables you can rate, files by drag-and-drop, a canvas each agent can draw on, and voice over the current chat. The same agents answer in Slack, Telegram and WhatsApp, on a phone call, over a public chat link, over MCP and from the CLI. Projects (Enterprise) give a cross-agent piece of work its tasks, its log, its health and a steward that keeps it moving.
- <table>
- <tr>
- <td width="50%"><a href="docs/screenshots/workspace-chat.webp"><img src="docs/screenshots/workspace-chat.webp" alt="Workspace: a chat with an ask card and the canvas rail"/></a><br/><sub><b>Workspace</b> - topic chats, an ask discussed in the thread, the rail open on the agent's canvas.</sub></td>
- <td width="50%"><a href="docs/screenshots/workspace-projects.webp"><img src="docs/screenshots/workspace-projects.webp" alt="Workspace Projects list"/></a><br/><sub><b>Projects</b> (Enterprise) - every project with its status, health and steward; tasks, a log and files inside.</sub></td>
- </tr>
- <tr>
- <td width="50%"><a href="docs/screenshots/workspace-project-detail.webp"><img src="docs/screenshots/workspace-project-detail.webp" alt="One Workspace project: overview, steward and health"/></a><br/><sub><b>One project</b> (Enterprise) - the goal, where it stands, the steward that keeps it moving, and its health.</sub></td>
- <td width="50%"><a href="docs/screenshots/workspace-start-chat.webp"><img src="docs/screenshots/workspace-start-chat.webp" alt="Start a chat with one or several agents" height="420"/></a><br/><sub><b>Several agents, one conversation</b> - pick the agents a chat should share.</sub></td>
- </tr>
- </table>
- ## The work compounds
- [Cornelius](https://github.com/Abilityai/cornelius) is our cognitive architecture for making the work compound. It turns what agents do into knowledge the company keeps: lessons from a run become notes, notes become instructions, instructions become the next agent's starting point - all as files in repositories you own, without retraining a model. Swap the model and what the company has built up stays.
- <p align="center">
- <a href="docs/screenshots/brain-orb.png"><img src="docs/screenshots/brain-orb.png" alt="The Brain Orb: a knowledge-base agent renders its own knowledge graph" width="720"/></a><br/>
- <sub><b>Brain Orb</b> (opt-in) - a knowledge-base agent renders its own notes and edges as a live graph.</sub>
- </p>
- ## Inside your perimeter. On any model.
- <!-- explainer graphic g3-laptop-to-company: brief §6.2, not yet produced -->
- Three runtimes - Claude Code, OpenAI Codex, Gemini CLI - and a model picker per chat. Each agent is its own container on a network the data plane cannot see. Credentials are encrypted at rest and scrubbed from the runtime; no telemetry leaves unless you opt in. Export every agent, every log and every credential, and walk away whenever you choose.
- <p align="center">
- <a href="docs/screenshots/model-picker.png"><img src="docs/screenshots/model-picker.png" alt="The per-chat model picker: Claude Code, Codex, Gemini CLI" width="720"/></a><br/>
- <sub><b>Model picker</b> - choose the runtime per chat; the agent's state, history and permissions do not change.</sub>
- </p>
- ## Quick start
- **One line, on a machine you control** (Docker and Docker Compose v2):
- ```bash
- curl -fsSL https://raw.githubusercontent.com/abilityai/trinity/main/install.sh | bash
- ```
- Then open `http://localhost`, claim the admin account, and connect a model in the setup overlay. Prefer to read what runs first: `git clone https://github.com/abilityai/trinity && cd trinity && ./scripts/deploy/start.sh` (`--hosted` pulls the prebuilt images; `--unattended` prints a generated admin password).
- **One click, in your own cloud account** - the image brings itself up behind HTTPS in about ninety seconds:
- [](https://marketplace.digitalocean.com/apps/trinity?refcode=d1f053807871&action=deploy) [](https://aws.amazon.com/marketplace/pp/prodview-q6f3biwa662o2)
- - **DigitalOcean** - [Marketplace](https://marketplace.digitalocean.com/apps/trinity?refcode=d1f053807871&action=deploy) or the guided [`trinity-do-create.sh`](https://docs.ability.ai/getting-started/deploying/digitalocean)
- - **AWS** - [Marketplace](https://aws.amazon.com/marketplace/pp/prodview-q6f3biwa662o2), or build the AMI and CloudFormation stack yourself with [packer/aws/README.md](packer/aws/README.md)
- - **Vultr** - [Marketplace image](scripts/deploy/vultr-marketplace.md)
- - **Any server you can SSH into** - `/trinity:deploy-new-instance` from the [abilities](https://github.com/abilityai/abilities) marketplace
- **Let your agent install it.** Tell Claude Code, Codex or any capable coding agent: *"Install Trinity on my computer. Follow the runbook at `https://raw.githubusercontent.com/abilityai/trinity/main/docs/AGENT_INSTALL_GUIDE.md`."*
- **Then deploy your first agent** from the coding agent you already use:
- ```bash
- /plugin marketplace add abilityai/abilities # Claude Code; Codex and Gemini CLI via the CLI below
- /plugin install trinity@abilityai
- /trinity:connect # instance URL + email code → MCP key
- /trinity:onboard # from any agent directory: check, package, deploy, start
- ```
- Or from a terminal: `pip install trinity-cli && trinity init && trinity deploy .` ([CLI reference](docs/CLI.md)). Or in the browser: **Create Agent** → a template, a GitHub repo, or a system manifest.
- > **An AI agent reading this?** [AGENTS.md](AGENTS.md) is your entry point: exact commands, key facts and a verification step for every path.
- > **Up and running? ⭐ [Star the repo](https://github.com/abilityai/trinity).** Hit a rough edge? [File an issue](https://github.com/abilityai/trinity/issues/new) - we read them.
- ## Open source is the whole platform. Enterprise adds governance.
- | **Open source - Apache 2.0, free** | **Enterprise adds** |
- |---|---|
- | **Agent platform** - fleet, Workspace, chat, voice, canvas, web terminal | **User & org management** - invite, deactivate, activity per user |
- | **Multi-runtime** - Claude Code, Gemini CLI, OpenAI Codex | **Single sign-on (OIDC)** - Okta, Entra ID, Google Workspace |
- | **Orchestration** - schedules, loops, agent-to-agent calls, fan-out, recovery | **Two-factor authentication** - authenticator apps, recovery codes, policy per role |
- | **Roles, objectives, gates** - role per agent, gated skills, approvers, Inbox | **Advanced RBAC** - teams, granular roles, people-per-agent assignments |
- | **System agent** - built-in ops agent: health, restarts, schedules, costs | **Projects** - tasks, log, health, steward across agents |
- | **Channels & APIs** - Slack, Telegram, WhatsApp, voice calls, MCP, REST, CLI | **Fleet permissions matrix** - one view of which agent may call which |
- | **Git-native agent state** - GitHub sync, commit history, reset to upstream | **Compliance suite** - audit dashboard, SIEM export, auditor export, retention |
- | **Skills library & marketplace** - tag-pinned library, skill sets, plugins | **Credential vault** - named encrypted secrets, deny-by-default grants |
- | **Security core** - encrypted credentials, hardened containers, audit log | **A2A management** - exposure control, inbound allowlist, paid calls |
- | **Guardrails & cost controls** - hooks agents cannot disable, quotas, loop budgets | **Cross-model validation** - a second model referees a run |
- | **Tamper-evident audit log** - opt-in SHA-256 hash chain | **Ephemeral agents** - short-lived agents with a lifetime and a budget |
- | **Self-hosted** - your hardware, no telemetry unless you opt in | **Licence management** - offline-verified, air-gap friendly |
- Open core: Enterprise includes everything in open source. The private modules live in optional submodules; cloning without them needs no credentials ([docs/ENTERPRISE.md](docs/ENTERPRISE.md)). [Get enterprise access →](mailto:hello@ability.ai)
- ## Build on it
- **A system from one manifest** - departments, agents, permissions and schedules, deployed together:
- ```yaml
- name: revenue
- agents:
- revenue-lead:
- template: github:abilityai/agent-corbin
- schedules: [{ name: weekly-pipeline, cron: "0 9 * * 1", message: "/pipeline-review" }]
- proposal-writer:
- template: github:abilityai/agent-ruby
- folders: { expose: true, consume: true }
- permissions:
- preset: full-mesh
- ```
- `deploy_system` over MCP, or `POST /api/systems/deploy`. [Multi-Agent System Guide →](docs/MULTI_AGENT_SYSTEM_GUIDE.md)
- **Templates and import** - start from the [template registry](https://github.com/abilityai/trinity-templates), any GitHub repo, or a local directory. **Skills library** - a tag-pinned community library ([trinity-skills](https://github.com/abilityai/trinity-skills)) plus your own repos; assign a skill or a set to any agent, re-injected fleet-wide when the library moves. **Plugins** - the [abilities](https://github.com/abilityai/abilities) marketplace covers scaffold, develop, deploy and iterate from Claude Code:
- | Plugin | What it does |
- |---|---|
- | **create-agent** | Scaffold a Trinity-compatible agent (`custom`, `website`, `clone`, `review`, `adjust`) |
- | **agent-dev** | Playbooks, memory systems, git sync, backlog workflow, pipelines, work loops |
- | **trinity** | `connect`, `onboard`, `sync`, `loop`, `create-dashboard`, `deploy-new-instance` |
- **MCP** - every Trinity instance is an MCP server (`http://<instance>:8080/mcp`, agent- and user-scoped keys) for agents, schedules, executions, projects, metrics, canvases, skills and more. No instance yet? `npx -y trinity-docs-mcp` answers questions about Trinity from any MCP client. [MCP reference →](docs/user-docs/integrations/mcp-server.md)
- ## Architecture
- <p align="center">
- <a href="docs/assets/trinity-deployment-topology.webp"><img src="docs/assets/trinity-deployment-topology.webp" alt="Deployment topology: one host, two isolated Docker networks, VPN-private operator access, tunnel-published public routes, PostgreSQL" width="720"/></a>
- </p>
- One host runs the control plane - a Vue frontend, a FastAPI backend, the MCP server, the scheduler, Redis for secrets and locks, and **PostgreSQL** for data - on a network the agents cannot reach. Each agent is its own container on a second network, with its repo, its credentials and its schedule. Operators reach the UI and API over your VPN; public users and channel webhooks reach only routes you publish through an outbound tunnel, so the host opens no inbound ports. Logs ship through Vector; metrics export over OpenTelemetry. Ports, services and profiles: [Deployment Guide](docs/DEPLOYMENT.md) · [docs/memory/architecture.md](docs/memory/architecture.md).
- > **Database.** Trinity 1.0 runs on PostgreSQL. The bundled container comes up with the stack; point `DATABASE_URL` at your own instance to use one you already run. SQLite support ended on 2026-09-01; an older instance migrates with the [Trinity Ops Agent](https://github.com/abilityai/trinity-ops-public)'s `/migrate-to-postgres`. <!-- engineering: compose default + docs follow the retirement issue -->
- ## We run our company on it
- Ability AI runs on Trinity: <!-- at tag: recount --> 41 agents across product, engineering, docs, operations and the company brain, as of 2026-10-14. Every number in this README is dated for that reason.
- **Security** - Grade A (Excellent) in an independent web application penetration test by [UnderDefense](https://underdefense.com), April 2026; all critical and high findings remediated. [Attestation letter →](docs/security/UnderDefense-Web-Pentest-Attestation-Apr-2026.pdf) · [SECURITY.md](SECURITY.md)
- ## Documentation
- - [**AGENTS.md**](AGENTS.md) - entry point for AI agents: task router, key facts, verification steps
- - [**User documentation**](docs/user-docs/README.md) - the Workspace, agents, roles and approvals, scheduling, channels, the API
- - [**Trinity Compatible Agent Guide**](docs/TRINITY_COMPATIBLE_AGENT_GUIDE.md) - what an agent repository needs to run here
- - [Deployment Guide](docs/DEPLOYMENT.md) · [PostgreSQL setup](docs/POSTGRESQL_SETUP.md) · [Versioning & upgrades](docs/VERSIONING_AND_UPGRADES.md)
- - [CLI reference](docs/CLI.md) · [MCP server](docs/user-docs/integrations/mcp-server.md) · [Multi-Agent System Guide](docs/MULTI_AGENT_SYSTEM_GUIDE.md)
- - [Abilities marketplace](https://github.com/abilityai/abilities) · [Skills library](https://github.com/abilityai/trinity-skills) · [Template registry](https://github.com/abilityai/trinity-templates) · [Trinity Ops Agent](https://github.com/abilityai/trinity-ops-public)
- - [Testing strategy](docs/testing/STRATEGY.md) · [Known issues](docs/KNOWN_ISSUES.md) · [Contributing](CONTRIBUTING.md)
- ## Community & support
- - **Ask Trinity**: [docs.ability.ai](https://docs.ability.ai) - the documentation, with an agent you can ask
- - **Issues**: [bugs and core maintenance](https://github.com/abilityai/trinity/issues) · **Discussions**: [ideas, questions, show and tell](https://github.com/abilityai/trinity/discussions)
- - **Workshops**: [live every Thursday, full archive](https://www.ability.ai/workshops) · **Discord**: [join](https://discord.gg/yQUqyp39)
- - **Security**: [SECURITY.md](SECURITY.md) · **Commercial**: [hello@ability.ai](mailto:hello@ability.ai)
- ## License
- [Apache License 2.0](LICENSE) - free for any use, commercial included, with an explicit patent grant. Enterprise modules are under a separate commercial licence. See [CONTRIBUTING.md](CONTRIBUTING.md) to contribute; development setup lives there now.
- ---
- <div align="center">
- <sub>Built by <a href="https://ability.ai">Ability AI</a> - the lab for AI-native companies</sub>
- </div>
Ratings and comments
Sign in to rate this skill.
Sign inFree to use. If it’s helped, pass it on.