████████╗██████╗ ██╗   ██╗███╗   ███╗██████╗  ██████╗ 
╚══██╔══╝██╔══██╗██║   ██║████╗ ████║██╔══██╗██╔═══██╗
   ██║   ██████╔╝██║   ██║██╔████╔██║██████╔╝██║   ██║
   ██║   ██╔══██╗██║   ██║██║╚██╔██║██╔══██╗██║   ██║
   ██║   ██║  ██║╚██████╔╝██║ ╚═╝ ██║██████╔╝╚██████╔╝
   ╚═╝   ╚═╝  ╚═╝ ╚═════╝ ╚═╝     ╚═╝╚═════╝  ╚═════╝ 

Trumbo

An open-source, self-hostable AI coding agent for your terminal.
Bring your own keys. No telemetry, no account, no upstream required.

Docs • Issues • Discussions

---

Trumbo is a self-hostable AI coding agent. It reads your project, plans a change, edits files across your codebase, runs shell commands, browses the web, and reports back — all with you in the loop. It runs two ways from one engine:

• CLI — an interactive terminal app plus a headless JSON mode for CI/CD and scripting.
• SDK — a TypeScript API for building your own agents, tools, connectors, and scheduled automations.

Everything is bring-your-own-key: plug in Anthropic, OpenAI, Google, OpenRouter, Bedrock, Vertex, Azure, Ollama, LM Studio, or any OpenAI-compatible endpoint. Trumbo never phones home — telemetry is disabled and there is no hosted backend to sign in to.

What's in this repo

Package	What it is	Path
CLI	Terminal UI, headless mode, connectors, schedules.	projects/console/
SDK	Programmatic agent engine, tools, plugins, cron.	engine/
Docs	Published documentation pages.	book/

Quick start

Install the CLI — pick any method. The published binary is self-contained, so no Node, Bun, or npm runtime is needed to run it.

# npm / pnpm / bun
npm i -g @trumbodev/cli
pnpm add -g @trumbodev/cli
bun add -g @trumbodev/cli

# curl (macOS / Linux) — no package manager required
curl -fsSL https://raw.githubusercontent.com/xedro98/trembo/main/projects/console/script/install.sh | sh

# PowerShell (Windows)
irm https://raw.githubusercontent.com/xedro98/trembo/main/projects/console/script/install.ps1 | iex

Then run it in your project:

cd my-project
trumbo

Or, from source:

git clone https://github.com/xedro98/trembo.git
cd trumbo
bun install
bun --conditions=development --cwd projects/console dev

It edits code across your project

Trumbo maps your project structure, understands how files relate, and makes coordinated edits across them. It watches linter and compiler output as it works and fixes missing imports, type mismatches, and syntax errors before you see them. Every edit is a diff you can review, tweak, or revert; checkpoints let you undo the agent's work in one step.

It runs your shell

Trumbo runs commands in your terminal and reads the output live — install deps, run builds, execute tests, deploy. Long-running processes (dev servers, watchers) keep running in the background and Trumbo reacts to new output, catching compile errors and test failures as they happen.

Plan and Act

Switch between Plan mode (explore, ask questions, lay out a strategy) and Act mode (execute the plan). Every file edit and command can require your approval, or flip on auto-approve to let it run autonomously.

Rules and skills

Drop project rules into .trumborules/ — coding standards, architecture conventions, deployment runbooks, testing requirements. They're picked up automatically by the CLI. Skills let the model load specific rule sets on demand.

Works with every model

Provider	Models
Anthropic	Claude Opus, Sonnet, Haiku
OpenAI	GPT series
Google	Gemini series
OpenRouter	200+ models from any provider
AWS Bedrock	Claude, Llama, and more
Azure / GCP Vertex	All hosted models
Cerebras / Groq	Fast inference
Ollama / LM Studio	Local models on your machine
Any OpenAI-compatible API	Self-hosted or third-party endpoints

Extend with plugins or MCP servers

Register tools and lifecycle hooks programmatically:

import { Agent, createTool } from "@trumbo/sdk"

const deployTool = createTool({
  name: "deploy",
  description: "Deploy the current branch to staging.",
  inputSchema: { type: "object", properties: { env: { type: "string" } }, required: ["env"] },
  execute: async (input) => {
    // your deployment logic
  },
})

const agent = new Agent({ tools: [deployTool] })

…or connect MCP servers for databases, APIs, cloud infra, and external systems. Manage them from the CLI with trumbo mcp.

Multi-agent teams

A coordinator agent splits work into subtasks and delegates to specialist agents, each with its own tools and context. Team state persists across sessions.

trumbo --team-name auth-sprint "Plan and implement user authentication with tests"

Scheduled agents

Run agents on cron schedules for recurring work — daily PR summaries, weekly dependency checks, health reports. Schedules persist across restarts and run independently of any terminal.

trumbo schedule create "PR summary" \
  --cron "0 9 * * MON-FRI" \
  --prompt "List all open PRs and their review status" \
  --workspace /path/to/repo

Connect to Slack, Telegram, Discord, and more

Chat with your agent from Telegram, Slack, Discord, Google Chat, WhatsApp, or Linear. Each thread maps to an agent session with full context.

trumbo connect telegram -k $BOT_TOKEN
trumbo connect slack --bot-token $SLACK_TOKEN --signing-secret $SECRET --base-url $URL

Headless CLI for CI/CD

Pipe input, get JSON out, chain commands, wire it into pipelines.

trumbo "Run tests and fix any failures"
git diff origin/main | trumbo "Review these changes for issues"
trumbo --json "List all TODO comments" | jq -r 'select(.type == "agent_event" and .event.text) | .event.text'

Contributing

Read the Contributing Guide. Open an issue or start a discussion if you want to help.

License

Apache 2.0 © 2026 Trumbo Bot Inc.