OpenGAP by Lyzr
Your repository becomes your agent
318 followers
Your repository becomes your agent
318 followers
Your AI agent's soul belongs in Git, not locked inside a framework. GitAgent is an open standard that extracts your agent's config, logic, tools, and memory into a portable, version-controlled definition. Define once. Run anywhere. Claude, OpenAI, CrewAI, OpenClaw, you name it. Same repo, any runtime. Roll back prompts like code. Branch, review, reproduce. #OwnYourAgent
Free Options
Launch Team
CloudflareYour next app belongs on Cloudflare Workers. Deploy for free
Promoted
congrats on the launch!
fun thing i stumbled upon while going through GitAgent : financial controls and git workflows are isomorphic, just with different names
maker-checker = pull request
sod (no single person can both create and approve a transaction) = branch protection
audit trail = git log
control documentation = rules.md.
so a git-native agent doesn't need to be "made compliant", it already is, just by how git works
pretty underrated consequence of this architecture for finance agents !
@priyaming Hey it would be great if you can raise an issue on github, we will look at it, thanks!
Hey Product Hunt! 👋
We're thrilled to share something we've been obsessing over. @GitAgent
Here's the backstory: We kept seeing the same problem over and over again. Teams would pour weeks into building AI agents, defining tool chains, decision logic, memory, persona, only to realize that all of that IP was trapped inside whatever framework they chose on day one.
Want to switch frameworks? Rewrite everything. Want to version-control your agent the way you version-control your code? No clean way to do it. Want to hand off your agent to another team or deploy it somewhere else? Pain.
We thought, why can't agents work like code? Store them in Git. Version them. Port them. Run them anywhere.
That's GitAgent.
It extracts the "soul" of your agent. The config, logic, tools, everything that makes it yours. It stores it as a portable, version-controlled definition. Then you deploy it to any framework with one command.
We're calling this #OwnYourAgents because we genuinely believe that if you built the agent, you should own it. Not the framework. Not the platform. You.
We'd love your feedback, your upvotes, and most importantly, your honest takes on what we can do better. This is just the beginning.
Let's make agent portability the standard. 🙌
The Git analogy genuinely clicks here — every problem you listed (no source of truth, no versioning, no portability) is exactly what Git solved for code in the 2000s, and it's wild that agents in 2025 are still in that pre-Git chaos.
The cross-runtime portability is the part I'd want to stress-test. Agents aren't just prompts — memory, tool access, and behavior can mean very different things in Claude vs CrewAI vs OpenClaw. Curious how GitAgent handles those runtime-specific differences without forcing you to lowest-common-denominator everything. Is there a layer that maps capabilities per runtime, or does the developer have to manage that themselves?
Also — branching and diffing agents sounds powerful but slightly terrifying. If I branch an agent's memory, what does merging that even look like?
This is the kind of infrastructure work that doesn't get enough attention because it's unglamorous. But if the standard gets adopted, everything else in the ecosystem gets easier. Worth watching closely.
The segregation of duties aspect is interesting. Are there ways that GitAgent helps enforce that or remind me to improve SOD when any one agent begins to take on too many steps within one process?
@lienchueh segregation of duties are provided by yaml file where in we can have that param.
The agent space has been circling this problem without quite naming it. Most tools compete at the runtime level, which means your agents are effectively hostage to whatever platform you chose first. GitAgent moves the leverage point to ownership, and that changes things.
The shift from agents as experiments to agents as assets is the real story here. Once they're versioned artifacts you can fork, audit, and evolve like code, they stop being fragile and start behaving like infrastructure.
That's also a sharper positioning angle than agents in Git. The deeper narrative is portability and permanence, something teams can actually own and build on. That lands hard with engineering led buyers who are already skeptical of lock-in.
I'm curious whether you see GitAgent staying a standard layer or becoming the primary interface for building and managing agents altogether.