- Man, it's such a shame that you pivoted away from the canvas-based editor concept, it was such a pleasure to use, it's so much better than tabs.
https://github.com/haystackeditor/haystack-editor
It's still probably the best tool to navigate, visualize and understand complex codebases, which is particularly important now with AI coded repos. I keep looking for alternatives but they are all notably worse.
About a month ago I spent a few frustrating hours building it from source for my system and making it work, and I've enjoyed using it as my main IDE since.
I wish I had the time to make a fork and bring in a newer version of VSCode. If anyone takes it up I might help at least.
- Yeah I agree it's a shame. Unfortunately coding has changed fast and I was not confident that the editor was in the correct direction with AI coding becoming so prevalent.
I think there is a lot of value with "reconnecting" with your codebase, so I do have some plans to bring the core concept of Haystack back in one form or another.
- 100% - tried forking VSCode too, scratched it. Way too much work to restyle, and I started wondering if IDEs should even be code-focused anymore. Built https://getmesa.dev instead, a canvas-based IDE for Claude Code & other CLI Agents. Kinda stuck on direction though - lmk if you've got ideas or wanna help out : )
- I'd certainly love to help! I've also been working on a canvas for building agentic flows and workspaces, and I'm actually at an intersection to what you're doing with that application you're showing off. I have a simpler node based stream system for flight control of arbitrary agent actions, meant to be operated by one or more orchestrating agents. Eg, the agent understands the surface it can create nodes with, generate dynamic tools and functions within the node specs, and then write graphs of them together so it's live and observable. Extremely intersectional, haha.
Happy to chat whenever you'd like, I can stalk a contact for ya or vice versa!
- Haystack is a popular Framework for building AI agents: https://haystack.deepset.ai/
- I like this idea. To be blunt, would this have more features than hooking up Claude/Gemini/Codex and saying "If - at any point - you're unsure, step back and let a human review"
- I think Haystack offers:
1. A centralized review mechanism for a team or org that operates on coding agent conversations in addition to diffs (and the codebase). It evaluates multiple different variables (e.g. how sensitive are the changes, how much did the author do to derisk, and what did the author's coding agent gloss over) and helps enforce your team's guidelines moreso than just an individual's prompt
2. Adversarial review that operates in addition to other AI review agents (e.g. BugBot, or Greptile) and filters any comments to only the things the author cares about. This helps cut down on the "AI reviewer battleground" that is present in pull requests
3. A review interface that allows human reviewers to quickly understand what the author did to verify their changes and focus on the author's design decisions
We actually jury-rigged all of this together before building Haystack, but found that it doesn't scale to the team level (since every individual has their own ideas/opinions of what constitutes a human review).
We also found that reviewing through purely Claude Code/Codex was slow and difficult because stuff like author traces are not pre-processed and you have to get your agent to specifically explore/understand them.
- Just to say great idea. But the name "Haystack" is used by several dozen things FWIW :)
33 points by akshaysg 1 day ago | 8 comments