Orbicue, tasks that know your codebase

Turn vague ideas into code-aware tasks

7 followers

Turn vague ideas into code-aware tasks

7 followers

Connect GitHub, index your codebase, and turn rough ideas into structured tasks with relevant files, subtasks, priorities, and labels. Keep work synced across Orbicue, GitHub Projects, your IDE, and MCP-powered coding agents.

Orbicue, tasks that know your codebase gallery image

Free

Launch tags:Productivity•Task Management•Software Engineering

Launch Team / Built With

Viktor.comAn AI coworker that actually does the work

Promoted

Maker

📌

Hey Product Hunt, I'm Taha, the maker of Orbicue. I built Orbicue around a gap I kept running into: issue trackers understand tickets, and coding assistants understand code, but the work of translating between the two still falls on the developer. A request often starts as something vague: "Clean up authentication." "Add team invitations." "Fix the onboarding flow." Before anyone can work on it, somebody has to inspect the repository, find the relevant files, understand the surrounding code, identify the edge cases, break the work into steps, create the ticket, and then keep the project board and GitHub updated afterward. Orbicue turns that handoff into one connected workflow. Connect a GitHub repository and index the codebase. Then describe the change in ordinary language. Orbicue retrieves relevant parts of the code and drafts an editable task with a clear title and description, priority and labels, an actionable checklist, and the files it used as context. Once the task exists, it stays connected to the rest of the development workflow. Orbicue includes two-way GitHub Projects sync, an extension for VS Code-compatible editors, and a built-in MCP server that lets coding agents read and update the same projects, boards, and tasks your team uses. Behind that is a complete project workspace with boards, lists, tables, calendars, timelines, sprints, docs, goals, analytics, comments, automations, and time tracking. The goal is not to let AI decide what should be built. The goal is to remove the repetitive translation work between a rough idea, the actual codebase, GitHub, the coding agent, and the project board. I would especially value feedback on the code-aware task generation: What context would make a generated engineering task genuinely useful to you, and what would you need to see before trusting the result?

Report

1mo ago

🧐 Good find

Honestly the MCP integration that syncs tasks back to GitHub Projects and the IDE at the same time is kind of brilliant, finally one source of truth that does not fall apart the second someone opens a PR.

Report

1mo ago

Maker

@zbeyde1568316 thank you for checking it out.

Report

1mo ago

Synced my repo and it actually pulled the right files for a fuzzy feature request, which is more than I expected. The GitHub Projects round-trip works clean too.

Report

1mo ago

Maker

@uur29784567603 thank you for the feedback.

Report

1mo ago

Hooked up my repo and asked for a quick task around our auth flow, the suggestion came back with the exact files involved and a clean breakdown of subtasks. That alone beats the manual ticket shuffle I usually do.

Report

1mo ago

Maker

@rza1192250 thank you, will ship more features soon.

Report

1mo ago

the way it pulls in the exact files and labels relevant subtasks right from a rough prompt feels really thoughtful, not just slapped together. nice execution.

Report

1mo ago

Maker

@sedat1276640 we are still trying to make the indexing better.

Report

1mo ago

Maker

Github features such as Project sync and Issues sync give us an Edge over other apps.

Report

1mo ago

Reviews