Contribution Guide

Overview

Contributions are welcome to Agent OS (sulala package) and to this documentation. This guide covers running from source, where to make changes, and how to submit contributions.

Running from source

Agent OS (sulala)

cd sulala
bun install
bun run cli onboard   # first time: create ~/.agent-os, seed agents/skills
bun run dev           # start server at http://127.0.0.1:3010

• CLI: bun run cli -- <command> [args] (e.g. bun run cli -- version, bun run cli -- run echo_agent "Hi").
• Build: bun run build produces dist/cli.js and dist/index.js.
• Tests / lint: Run any existing test or lint scripts in the sulala package (e.g. bun test, or as documented in the repo).

Documentation (this site)

cd docs
npm install
npm run dev

Edit Markdown under docs/content/. Structure: 1.getting-started/, 2.guides/, 3.platform/, 4.troubleshooting/. The site uses the project’s docs stack (e.g. Docus/Nuxt Content).

Where to make changes

• Core: sulala/src/core/ — config, agent-registry, runtime, llm, tool-registry, tasks, events, graphs, plugins.
• HTTP: sulala/src/http/ — handlers, channel webhooks (telegram, slack, discord, signal, viber), memory, conversations.
• Skills: sulala/src/skills/loader.ts — skill and tools loading.
• Tools: sulala/src/tools/ — built-in tools (echo, time, exec, run-agent).
• CLI: sulala/src/cli.ts — commands (version, start, stop, onboard, update, run).
• Docs: docs/content/ — all user-facing docs; keep them aligned with the Agent OS codebase and roadmap (sulala/roadmap.md).

Submitting contributions

• Open a pull request (or follow the repo’s process). For docs, ensure links and code samples match the current Agent OS (Bun, port 3010, ~/.agent-os, sulala CLI commands).
• For skills, add or update skills in data/skills/ and document any new config or tools in Skills authoring.