This product was not featured by Product Hunt yet. It will not be visible on their landing page and won't be ranked (cannot win product of the day regardless of upvotes).
DevWit is a free, open-source AI IDE built on one principle: you should see exactly what the AI sees and does. Every LLM request shows its full context — system prompt, tools, injected code — with per-item token costs you can toggle. Every file write and shell command needs your explicit approval.
Hi Product Hunt! I'm the maker of DevWit.
I built it because I kept hitting the same wall with AI coding tools: I couldn't tell what was being sent to the model, or what the agent was about to do to my files. DevWit makes both fully visible:
1. Context Panel — every LLM request lists its system prompt, tools, and each injected item (file, RAG chunk, terminal output) with token counts. Toggle any item off and the request shrinks accordingly.
2. Authorization Gate — the agent cannot write a file or run a shell command without your one-click approval. Approvals can be remembered per-project.
3. Lean context — a task router sends simple requests to local models (Ollama) and complex ones to cloud models, with a per-session cost ledger.
It's a real IDE, not a wrapper: TypeScript LSP (diagnostics/hover/definition), integrated Git, breakpoint debugging (DAP/js-debug), MCP tool servers, multi-agent orchestration, custom modes — all free, all open source.
v0.3.0 ships Windows/macOS/Linux builds verified by 614 unit tests and 28 end-to-end suites. No accounts, no cloud sync, telemetry is opt-in and off by default.
I'd love your feedback — especially from teams with compliance/audit needs.
The token cost toggle per item is such a thoughtful touch, makes it way easier to figure out where the real spend is happening. Love that you kept it open source.
The per-item token cost toggle is a clever touch, makes it way easier to understand where your context window is actually going.
Thanks @Furkan! Really glad the token cost breakdown resonated — it was one of those features that seemed obvious once built but nobody else was doing it. The approval gate came from real pain: an agent once overwrote a file I didn't expect, and I thought 'never again.' Appreciate you checking it out!
Thanks @Report! Appreciate you checking out DevWit. If you have any questions or feedback, I'm happy to help.
Thanks @Report! Appreciate you checking out DevWit. If you have any questions or feedback, I'm happy to help.
Thanks @Report! Appreciate you checking out DevWit. If you have any questions or feedback, I'm happy to help.
Hi Product Hunt! I'm the maker of DevWit.
I built it because I kept hitting the same wall with AI coding tools: I couldn't tell what was being sent to the model, or what the agent was about to do to my files. DevWit makes both fully visible:
1. Context Panel — every LLM request lists its system prompt, tools, and each injected item (file, RAG chunk, terminal output) with token counts. Toggle any item off and the request shrinks accordingly.
2. Authorization Gate — the agent cannot write a file or run a shell command without your one-click approval. Approvals can be remembered per-project.
3. Lean context — a task router sends simple requests to local models (Ollama) and complex ones to cloud models, with a per-session cost ledger.
It's a real IDE, not a wrapper: TypeScript LSP (diagnostics/hover/definition), integrated Git, breakpoint debugging (DAP/js-debug), MCP tool servers, multi-agent orchestration, custom modes — all free, all open source.
v0.3.0 ships Windows/macOS/Linux builds verified by 618 unit tests and 28 end-to-end suites. No accounts, no cloud sync, telemetry is opt-in and off by default.
I'd love your feedback — especially from teams with compliance/audit needs.
The token cost breakdown per request is honestly really useful, kind of eye opening when you see where the budget actually goes. Love that the approval step is baked in by default.
About DevWit on Product Hunt
“The AI IDE that shows you exactly what it sends”
DevWit was submitted on Product Hunt and earned 5 upvotes and 9 comments, placing #110 on the daily leaderboard. DevWit is a free, open-source AI IDE built on one principle: you should see exactly what the AI sees and does. Every LLM request shows its full context — system prompt, tools, injected code — with per-item token costs you can toggle. Every file write and shell command needs your explicit approval.
DevWit was featured in Open Source (68.7k followers), Developer Tools (517.5k followers), Artificial Intelligence (475.9k followers) and GitHub (41.4k followers) on Product Hunt. Together, these topics include over 233.9k products, making this a competitive space to launch in.
Who hunted DevWit?
DevWit was hunted by ZONGYUAN SUI. A “hunter” on Product Hunt is the community member who submits a product to the platform — uploading the images, the link, and tagging the makers behind it. Hunters typically write the first comment explaining why a product is worth attention, and their followers are notified the moment they post. Around 79% of featured launches on Product Hunt are self-hunted by their makers, but a well-known hunter still acts as a signal of quality to the rest of the community. See the full all-time top hunters leaderboard to discover who is shaping the Product Hunt ecosystem.
Want to see how DevWit stacked up against nearby launches in real time? Check out the live launch dashboard for upvote speed charts, proximity comparisons, and more analytics.
ZONGYUAN SUI