Project Clyde Is Live. Download It Now.
Project Clyde Is Live. Download It Now.
It's here.
After three days of building the case, telling the story, and showing you what's possible, Project Clyde is now open for everyone to use. The code is live. Everything you've read over the past 72 hours is real, it's yours, and it's ready to change how you work with AI agents.
The Journey to Here
Over the last three days, we walked you through the problem: AI agents are powerful, but they're scattered. A single agent can do interesting things, but the moment you need multiple agents working together with consistent quality across complex workflows, friction explodes. We introduced our philosophy: the future isn't about creating the perfect singular agent, but about building thoughtful, specialized teams that grow smarter over time. We showed you what that looks like in practice: agents with distinct roles, reusable skills, persistent memory, and the ability to orchestrate themselves.
Today, you don't have to imagine anymore. You can build it.
What You're Getting
Project Clyde gives you everything you need to move beyond single-agent workflows.
Agent Registry — Create specialist agents with custom roles, system prompts, and assigned tools. Define exactly what each agent does, how they think, and what capabilities they have.
Skills Management — Build reusable process documents that encode how work gets done consistently. Assign skills to agents. Every agent follows the same best practices and proven approaches. Skills are how you scale expertise.
Agent Memory — Persistent knowledge that accumulates across tasks and projects. Agents don't start from zero each time. They learn from previous work, retain context, and compound their effectiveness over time.
Scheduled Tasks — Set agents to work automatically on recurring schedules using cron-style syntax. Automate the predictable work, freeing your team to focus on what actually matters.
File Triggers — Watch directories and automatically trigger agent actions when files appear or change. Drop a file, watch your agents respond.
Performance Tracking — Monitor agent quality, log feedback, and review performance history. Make decisions based on data, not guesses.
Self-Improvement — Agents can have their system prompts refined based on performance data. Your team doesn't just work. It improves.
Team Orchestration — Agents can spawn sub-teams for complex parallel work. Break large problems into specialized teams, coordinate their efforts, and combine results.
Built on Claude Code — Every agent is powered by Claude, so you're working with the latest in reasoning, code execution, and multimodal capabilities.
How to Get Started
Download the project at https://github.com/noeltg77/project-clyde. You'll get the full codebase, documentation, and everything needed to run agents locally or in your own environment.
Create your first agent using the Agent Registry. Define a role, write a system prompt, assign tools.
Explore the examples included in the repository. We've built sample agents and workflows to show you what's possible.
Assign a skill to your agent. Take one of the included process documents and attach it.
Run a task and review the results. See the agent in action. Check the performance tracking to understand what happened.
From there, you're building. Add another agent. Connect them. Set up a schedule. Watch your team work.
Who This Is For
Project Clyde is built for builders and experimenters. It's for teams managing multiple AI workflows who want them coordinated, not scattered. It's for anyone who's hit the ceiling of what a single agent can do and realized they need a team.
You might be:
- A researcher exploring multi-agent systems
- A startup automating internal processes
- A team lead trying to scale without hiring more people
- An engineer building the next generation of AI workflows
- Someone who just wants to see what's possible
If you've thought "I need more than one agent working together, learning from each other" — this is for you.
Community and Support
Project Clyde has a growing community of builders and experimenters.
Join the community at https://www.patreon.com/cw/ProjectClyde. Share what you're building, ask questions, and learn from others doing similar work. The roadmap is shaped by what this community needs.
Read the full documentation included with the download. We've written detailed guides on agent design, skill creation, memory management, and orchestration patterns.
Share your feedback directly with the team. What works? What's missing? What would unlock the next level of possibility? We're listening, and your input shapes what we build next.
This Is Just the Beginning
Project Clyde is built on an architecture designed to evolve. Today, it's what you see. Tomorrow, it's what you and the community build it into.
The core principles won't change: specialist agents, consistent quality through shared skills, persistent growth through memory, and intelligent orchestration. The specific features and workflow patterns will expand as the community explores what's possible.
You're not downloading a finished product. You're joining others building the next generation of AI work.
Download and Build
Project Clyde is live. Everything you need to get started is waiting for you.
Download now: https://github.com/noeltg77/project-clyde
Read the docs. Build your first agent. Start orchestrating your team. Join thousands of others exploring what's possible when AI agents work together.
We started this week by validating a problem. We told you why we built this. We showed you what it can do. Now it's your turn.
Thank you for following along over these three days. Thank you for believing in what we're creating. Thank you for being willing to explore the future with us.
The best work is still ahead. Let's build it together.
Get started: Download at https://github.com/noeltg77/project-clyde. Have questions? Join the community: https://www.patreon.com/cw/ProjectClyde