Error monitoring that opens the fix PR, not just the alert
Send an event in minutes, triage from the keyboard, and let the Conductor turn the error into a draft pull request you review and merge. Sentry-SDK compatible, so you can start by changing a DSN.
You do not need another dashboard to babysit. Condux is developer-first: fast issue list, real stack traces linked to the exact code, and an AI fix engine that does the boring first draft of the fix as a PR you own.
Start in minutes
First-party SDKs, or keep the Sentry SDK you already use.
import { init, captureException } from "@condux/node";init({ dsn: process.env.CONDUX_DSN });try {doWork();} catch (error) {captureException(error);throw error;}
Triage without leaving the keyboard
- j / k to move through issues
The list is built for working a backlog down fast.
- e / i / a to resolve, ignore, assign
Triage the open issue in one keystroke.
- Code-path stack traces
Frames link to the exact line on GitHub via your code mappings.
- An MCP endpoint
Point your AI agent at a project to query issues over the Model Context Protocol.
Let the Conductor draft the fix
Pick an issue and the Conductor assembles a scoped, double-scrubbed context, sends it to Claude or your own model, and opens a draft pull request on a new branch. You review and merge it through your own CI. It is never an auto-merge, and the model never sees your repo token.
Frequently asked questions
Do I have to learn a new SDK?
No. Condux is Sentry-SDK compatible, so you can point your existing SDK at it by changing the DSN, or use the first-party @condux/node and Python SDKs.
Is the AI fix an auto-merge?
Never. Every fix is a draft pull request your team reviews and merges through your own CI. Condux opens it; your team owns it.