May be some grouping.
Well, I got 30 defects and many of them looked like symptoms of one data sync problem. It could be useful to see related findings grouped by likely cause, so 30 rows don’t look like 30 separate fixes.
Also one High finding about sign-in missed that my app already has a recovery path, so a note on what the test could not check would help (but this is really a minor thing).
And maybe, it make sense also to add a "Cancel" option for the Run in phase when the system reads the code. Just to protects your resources if the run was started accidently by mistake.
Hey Product Hunt!!! I’m Malo, co-founder of heal.dev.
The problem:
👉 Code review and static analysis often miss real runtime bugs and vulnerabilities.
👉 Even with top-performing coding agents, we still ship bugs and vulnerabilities to production. We’ve seen that firsthand by running Heal on a collection of open-source repositories: https://www.heal.dev/#sandboxes.
The solution
👉 Instead of just scanning your code, Heal spins up a production-like sandbox with your external dependencies mocked.
👉 Heal finds bugs, perf issues, and vulnerabilities at runtime. It reproduces each against the sandbox before reporting.
👉 Unlike asking a coding agent to write end-to-end tests, Heal proactively hunts for edge cases and runs exploratory testing, not just the happy path.
The key difference
👉 Review bots find what a developer would change to keep code clean. A lot of findings.
👉 Heal finds a proven user behaviour that causes a bug. Fewer findings, closer to the user.
With Heal, you get high-signal, verified bugs and vulnerabilities users would see, and a completely autonomous system. It's free during the beta, no cc required: www.heal.dev
Debate club 🥊 Ding ding ding!
One question for y'all: what makes it so that you don't auto-merge all PRs? Could Heal be the reason you finally do?
@therealmalo 💪🏻
Was waiting for something like this!
Thanks @mikestaub and don't hesitate to share any harsh feedback :)
This found things we didn't expect on our codebase. Still wondering how to have it run fully in an automated loop -> part of our CICD process and weekly quality runs over the weekend, and when something is found a PR gets opened automatically. Great job @therealmalo @pierre_oucif1 @alexbeno !
Thanks @amaury_soviche1 ! Working on it :)
What you can do is let your coding agent query Heal via MCP, and let it handle the defects founds, then re-run to validate.
We're working on speed and UX on how this can fit neatly in a meta-harness.
We asked ourselves how Heal compares to AI code reviewers, and whether stacking two or three of them closes the gap. So we built a benchmark to find out.
TL;DR:
70% of the defects Heal finds, no code reviewer catches. Heal also only flags defects that are on a real user path
Stacking reviewers doesn’t help much: they overlap heavily on the same classes of bugs and all miss the ones that only show up when the app actually runs.
Full methodology and the defect list: https://www.heal.dev/benchmark
@bengeekly That's out of our current scope, but it brings about some ideas for the future. Currently, we hunt for issues in the application code, and we don't run it on libs. Some topic for us to investigate!
@bengeekly the two CVEs are in the php interpreter itself and our agent is focusing on bugs in the codebase, not inside the installed libs, language api, etc. But as these CVEs are related to memory management and so runtime, it's a bit related/adjacent to the runtime bugs we are finding to reproduce user (human and machine) behaviors.
I love the idea! However, my company uses CodeCommit for most of our projects rather than GitHub. Are there any plans in the future to add local Git compatibility?
@tj_ferry hey! We already have a local version of our agent that we are using for R&D. We plan to release it in couple of months ;)
Ran Heal against our own app and it found real bugs, each with repro steps and a test to prove it. A stored XSS and a cross-tenant data leak were the big ones. The API meant we could triage all 239 findings in about an hour. Congrats on the launch @therealmalo and @pierre_oucif1 .
Thanks @mark_ridgeon1 ! We're very grateful for all your feedback along the way!!