Launching today
AgentRail
A local control plane for AI coding agents
51 followers
A local control plane for AI coding agents
51 followers
AgentRail is the open-source task lifecycle layer for coding agents. Agents can write code β the hard part is everything else: picking the right ticket, watching CI, incorporating review feedback, merging when it's ready. It connects agents to GitHub, Linear, and CI with structured typed events instead of raw webhooks and log walls. - Per-agent scoped auth - Rules-based task-> agent routing, - structured CI summaries. - Apache 2.0 licensed. Runs locally.
AgentRail
Hey Product Hunt π
I'm Onyeka, and I've been obsessed with one problem: coding agents can write great code, but they can't close a ticket.
They miss CI failures. They don't know which issue to pick up next. They lose context on review comments. You end up babysitting them instead of letting them ship.
That's after spending a couple hours of your day setting this all up.
**The problem** π
Every agent integration I saw was the same: dump GitHub webhooks at your agent and hope for the best. Raw log walls. No structure. No routing. The agent burns context just parsing noise.
What we built β‘
AgentRail is an open-source task lifecycle API that gives coding agents a real dev loop:
- π₯ GitHub Issues and Linear tickets become routable tasks automatically
- π CI results come back as structured summaries β not raw log walls
- π PR review comments are ranked by severity (blocking vs. advisory)
- π Per-agent scoped API keys β each agent only touches what it needs
- π Runs entirely locally, no account required
**How to try it** π
npm install @agentrail-core/cli β that's it.
Completey free for self-hosted, contributions are more than welcome!
Agentrail is in the very early stages, and actually launched earlier than planned to make it for Vercel Day! So please, break it, roast the hell out of it, all feedback is welcome/
Would love to hear from anyone running Claude Code, Codex, or Cursor agents at scale. What's the first thing that breaks when you have more than one task running? π
A local control plane for coding agents is a huge win for security-conscious teams. Does AgentRail support fine-grained permissions for what files an agent can actually modify?
AgentRail
Honestly, the babysitting part is the biggest bottleneck with coding agents right now. Most tools just dump webhooks and hope for the best. I am curious about how does the structured CI summary handle complex logs from something like a massive monorepo without losing the specific error context for the agent?
AgentRail
@ritikgupta_01Β Hey Ritik, great question. We donβt summarize CI as one giant blob. AgentRail keeps the CI hierarchy: workflow β job β step/shard/package/test. The agent gets the failing leaves, key error excerpts, affected files/tests when available, commit/check IDs, and links back to the raw logs. So the context is compact, but the exact failure remains traceable instead of being lost in a generic summary.
HasData
the webhook β raw json β custom parser loop is the part of agent infra i hate most. how heavy is the setup?
AgentRail
@sergeynagornyΒ Very intentionally not heavy. You run agentrail init, connect a provider and then create an agent. Webhooks are optional; local setups can use polling first (for webhooks you'll need either a VPS or ngrok, since webhooks won't work on localhost). AgentRail handles provider payloads, signature checks, idempotency, routing, and normalized task state so agents donβt need bespoke raw JSON parsers.