[GH-ISSUE #358] Open to collaborators?

kerem commented

2026-03-03 13:52:45 +03:00

Owner

Originally created by @0xdevalias on GitHub (Mar 4, 2025).
Original GitHub issue: https://github.com/jehna/humanify/issues/358

Hey @jehna, was wondering if you were open to potentially having me join the humanify project as a more 'official' collaborator? Totally understandable if you'd prefer not, but I was thinking it would allow me to help triage/issues better/etc.

I also have a few ideas/local WIP refactors to clean up some of the CI stuff, and then maybe also refactor some parts of the codebase + add new features/etc (which I'll probably do as PR's regardless of the outcome of this)

No rush/pressure, but let me know what you think when you get a chance :)

Originally created by @0xdevalias on GitHub (Mar 4, 2025). Original GitHub issue: https://github.com/jehna/humanify/issues/358 Hey @jehna, was wondering if you were open to potentially having me join the `humanify` project as a more 'official' collaborator? Totally understandable if you'd prefer not, but I was thinking it would allow me to help triage/issues better/etc. I also have a few ideas/local WIP refactors to clean up some of the CI stuff, and then maybe also refactor some parts of the codebase + add new features/etc (which I'll probably do as PR's regardless of the outcome of this) No rush/pressure, but let me know what you think when you get a chance :)

kerem commented

2026-03-03 13:52:45 +03:00

Author

Owner

@0xdevalias commented on GitHub (Apr 21, 2025):

Some notes/context from a recent email conversation, RE: my interest in potentially forking humanify to be able to more easily work on things (but not quite having the capacity in my schedule to focus on it just yet):

Yeah, I similarly sent them a LI request a little while back, opened an issue about collaborating, and even more recently tried a direct email; and no response from any of them. They did end up merging a couple of my tiny PRs back in early February (1, 2), but not this slightly more involved/untested one in late February; and haven't seen anything from them since. From a quick skim across their blog/socials/etc it seems they've been pretty absent from there for quite a while now too. They've still been somewhat active on GitHub though according to their contributions, mostly in private repos, but a couple of public ones too (1, 2). This was the email I sent a couple of days ago:

Heya,

Forgive the direct contact; this will be the only email I send unless I hear back from you positively otherwise.

I noticed that you haven't been very active on the humanify repo for a while now; and from a quick skim of socials/etc maybe not there either; and so I wanted to reach out and see if you were open to adding collaborators to the humanify project?

https://github.com/jehna/humanify/issues/358

I have a number of ideas of refinements/improvements that I'd love to dive deeper into (as soon as I clear my current little backlog of unrelated tasks); but I'll admit that the lack of activity on the repo has made me a little hesitant, as I don't want to head in a direction you wouldn't be comfortable with, or have the PRs just sit there unmerged/similar.

Anyways, hope things are well with you; and sorry again for the direct reachout.

Cheers

Glenn 'devalias' Grant

RE: forking; yeah, that's basically been in my mind for a while now, though I have been pretty busy with other life things (travel, catching up on all the things I ignored while travelling, etc) so hadn't got around to it yet. In reality I will likely be somewhat busy for the next few weeks or so while I get back on top of everything, so I wouldn't be able to put too much focus into this till that is done; but I am definitely interested in exploring the idea of forking the project and working to get it cleaned up a bit more + get some new features/etc landed. (There's also another semi-related side project I have in mind that I hope to be able to put some more solid time into once I get on top of everything else too; but that is far more at an early R&D sort of level of things for now)

And then this followup:

this love is why you should be co-maintainer or just do a fork

Yeah, I'd be happy to just help out on the current repo if @jehna was open to that; but given the lack of contact/etc, fork is probably the next best thing.

@0xdevalias commented on GitHub (Apr 21, 2025): Some notes/context from a recent email conversation, RE: my interest in potentially forking `humanify` to be able to more easily work on things (but not quite having the capacity in my schedule to focus on it just yet): > Yeah, I similarly sent them a LI request a little while back, opened an [issue about collaborating](https://github.com/jehna/humanify/issues/358), and even more recently tried a direct email; and no response from any of them. They did end up merging a couple of my tiny PRs back in early February ([1](https://github.com/jehna/humanify/pull/309), [2](https://github.com/jehna/humanify/pull/308)), but not [this](https://github.com/jehna/humanify/pull/346) slightly more involved/untested one in late February; and haven't seen anything from them since. From a quick skim across their blog/socials/etc it seems they've been pretty absent from there for quite a while now too. They've still been somewhat active on GitHub though [according to their contributions](https://github.com/jehna), mostly in private repos, but a couple of public ones too ([1](https://github.com/jehna/grillmaster), [2](https://github.com/jehna/my-kali-setup)). This was the email I sent a couple of days ago: > > > Heya, > > > > Forgive the direct contact; this will be the only email I send unless I hear back from you positively otherwise. > > > > I noticed that you haven't been very active on the humanify repo for a while now; and from a quick skim of socials/etc maybe not there either; and so I wanted to reach out and see if you were open to adding collaborators to the humanify project? > > > > - https://github.com/jehna/humanify/issues/358 > > > > I have a number of ideas of refinements/improvements that I'd love to dive deeper into (as soon as I clear my current little backlog of unrelated tasks); but I'll admit that the lack of activity on the repo has made me a little hesitant, as I don't want to head in a direction you wouldn't be comfortable with, or have the PRs just sit there unmerged/similar. > > > > Anyways, hope things are well with you; and sorry again for the direct reachout. > > > > Cheers > > > > Glenn 'devalias' Grant > > **RE: forking;** yeah, that's basically been in my mind for a while now, though I have been pretty busy with other life things (travel, catching up on all the things I ignored while travelling, etc) so hadn't got around to it yet. In reality I will likely be somewhat busy for the next few weeks or so while I get back on top of everything, so I wouldn't be able to put too much focus into this till that is done; but I am definitely interested in exploring the idea of forking the project and working to get it cleaned up a bit more + get some new features/etc landed. (There's also another semi-related side project I have in mind that I hope to be able to put some more solid time into once I get on top of everything else too; but that is far more at an early R&D sort of level of things for now) And then this followup: > > this love is why you should be co-maintainer or just do a fork > > Yeah, I'd be happy to just help out on the current repo if @jehna was open to that; but given the lack of contact/etc, fork is probably the next best thing.

kerem commented

2026-03-03 13:52:46 +03:00

Author

Owner

@skilbjo commented on GitHub (Apr 21, 2025):

+1 allowing @0xdevalias to co-maintain this project w @jehna , or @0xdevalias if you want to start the fork, i will co-maintain with you.

assuming neither, i will just do a hard-fork myself, this project is too cool and exciting to leave unmaintained.

@skilbjo commented on GitHub (Apr 21, 2025): +1 allowing @0xdevalias to co-maintain this project w @jehna , or @0xdevalias if you want to start the fork, i will co-maintain with you. assuming neither, i will just do a hard-fork myself, this project is too cool and exciting to leave unmaintained.

kerem commented

2026-03-03 13:52:46 +03:00

Author

Owner

@0xdevalias commented on GitHub (Apr 21, 2025):

if you want to start the fork, i will co-maintain with you

@skilbjo I'm open to doing this, though might not have the capacity to do so immediately.

As I raised in my original email, did you have thoughts about the structure of the fork (eg. as an organisation, etc?):

Interested to hear your thoughts on future directions for the project. Are you interested in collaborating/working on the Humanify fork with me? Or more just hoping to see the project get some more direct love/updates? Did you have any particular thoughts on how you would envision the fork? Like in a new GitHub org for it, or?

Also, if we are going to fork, I'd also be interested to hear if @j4k0xb (https://github.com/j4k0xb/webcrack) / @pionxzh (https://github.com/pionxzh/wakaru) have any interest in being involved in it in some capacity (either directly at a co-maintainer sort of level; or more casually/indirectly); as I suspect there is some crossover potential with webcrack in particular (that humanify builds upon); and it's upcoming plugins feature/etc.

I suspect there is probably some re-architecting that could be done to make humanify nicer UX / DX overall (like it's always mildly bugged me the way the CLI splits local / remote LLMs / model downloads across separate commands/etc; as one example)

@0xdevalias commented on GitHub (Apr 21, 2025): > if you want to start the fork, i will co-maintain with you @skilbjo I'm open to doing this, though might not have the capacity to do so immediately. As I raised in my original email, did you have thoughts about the structure of the fork (eg. as an organisation, etc?): > Interested to hear your thoughts on future directions for the project. Are you interested in collaborating/working on the Humanify fork with me? Or more just hoping to see the project get some more direct love/updates? Did you have any particular thoughts on how you would envision the fork? Like in a new GitHub org for it, or? --- Also, if we are going to fork, I'd also be interested to hear if @j4k0xb (https://github.com/j4k0xb/webcrack) / @pionxzh (https://github.com/pionxzh/wakaru) have any interest in being involved in it in some capacity (either directly at a co-maintainer sort of level; or more casually/indirectly); as I suspect there is some crossover potential with `webcrack` in particular (that `humanify` builds upon); and it's upcoming plugins feature/etc. I suspect there is probably some re-architecting that could be done to make `humanify` nicer UX / DX overall (like it's always mildly bugged me the way the CLI splits local / remote LLMs / model downloads across separate commands/etc; as one example)

kerem commented

2026-03-03 13:52:46 +03:00

Author

Owner

@skilbjo commented on GitHub (Apr 21, 2025):

yea, if i hardfork it i would redo the project organisation in my favourite style, similar to what i've done in https://github.com/xhrdev/examples

ie

src/ folder
test/ folder
use jest
use the eslint settings i am familiar with
use a Makefile

i don't have any opinions on nicer UX/DX. i think you would be perfectly suited to make those improvements. i just see such good PRs being left unmerged, and it bothers me.

@skilbjo commented on GitHub (Apr 21, 2025): yea, if i hardfork it i would redo the project organisation in my favourite style, similar to what i've done in https://github.com/xhrdev/examples ie `src/` folder `test/` folder use `jest` use the `eslint` settings i am familiar with use a `Makefile` i don't have any opinions on nicer UX/DX. i think you would be perfectly suited to make those improvements. i just see such good PRs being left unmerged, and it bothers me.

kerem commented

2026-03-03 13:52:46 +03:00

Author

Owner

@0xdevalias commented on GitHub (Apr 21, 2025):

@skilbjo Hrmm.. fair enough. Makefile is somewhat non-canonical for a node project IMO.

With forking it, I was thinking of making a new org and having the repo in that; potentially as something that could act as collection of web reversing tools like this.

i just see such good PRs being left unmerged, and it bothers me.

@skilbjo Understandable.

@0xdevalias commented on GitHub (Apr 21, 2025): @skilbjo Hrmm.. fair enough. `Makefile` is somewhat non-canonical for a node project IMO. With forking it, I was thinking of making a new org and having the repo in that; potentially as something that could act as collection of web reversing tools like this. --- > i just see such good PRs being left unmerged, and it bothers me. @skilbjo Understandable.

kerem commented

2026-03-03 13:52:46 +03:00

Author

Owner

@brianjenkins94 commented on GitHub (Jun 2, 2025):

Just wondering if anything came of this? I would really like to have better handling of name conflicts.

https://github.com/jehna/humanify/issues/330

@brianjenkins94 commented on GitHub (Jun 2, 2025): Just wondering if anything came of this? I would really like to have better handling of name conflicts. https://github.com/jehna/humanify/issues/330

kerem commented

2026-03-03 13:52:46 +03:00

Author

Owner

@0xdevalias commented on GitHub (Jun 2, 2025):

Just wondering if anything came of this?

@brianjenkins94 The short answer is that despite trying to contact them in multiple ways over time, and despite them having recent GitHub activity, there has been no response from @jehna about this at all.

Based on that, @skilbjo and I have been talking about forking the project. Their plan was to hardfork it with a new name; I probably would have just forked it as something like humanify-ng and kept it closer to the ability to merge the projects back together in future if @jehna ever comes back to us on this.

But so far, to my knowledge / last time I checked in, I don't believe @skilbjo had been working on the fork so much due to other commitments; and similarly, I unfortunately haven't yet had capacity to look deeper into it myself.

@0xdevalias commented on GitHub (Jun 2, 2025): > Just wondering if anything came of this? @brianjenkins94 The short answer is that despite trying to contact them in multiple ways over time, and despite them [having recent GitHub activity](https://github.com/jehna?tab=overview&from=2025-04-01&to=2025-04-30), there has been no response from @jehna about this at all. Based on that, @skilbjo and I have been talking about forking the project. Their plan was to hardfork it with a new name; I probably would have just forked it as something like `humanify-ng` and kept it closer to the ability to merge the projects back together in future if @jehna ever comes back to us on this. But so far, to my knowledge / last time I checked in, I don't believe @skilbjo had been working on the fork so much due to other commitments; and similarly, I unfortunately haven't yet had capacity to look deeper into it myself.

kerem commented

2026-03-03 13:52:46 +03:00

Author

Owner

@skilbjo commented on GitHub (Jun 2, 2025):

yup @ that's the current status !

both @0xdevalias and i are hoping @jehna returns and just maintains this project, but in case he doesn't, i am keen to build a very similar project (and mostly inspired by) humanify, that i am calling Cipher. i think @0xdevalias would be a great co-maintainer on this with me, but i want to get it a little more built out before i invite him to the project.

i was able to get mostly what i was looking for out with aistudio.google.com in the meantime tho, hence why it hasn't been much of a priority for me. however i still am keen to do it!

@skilbjo commented on GitHub (Jun 2, 2025): yup @ that's the current status ! both @0xdevalias and i are hoping @jehna returns and just maintains this project, but in case he doesn't, i am keen to build a very similar project (and mostly inspired by) humanify, that i am calling Cipher. i think @0xdevalias would be a great co-maintainer on this with me, but i want to get it a little more built out before i invite him to the project. i was able to get mostly what i was looking for out with aistudio.google.com in the meantime tho, hence why it hasn't been much of a priority for me. however i still am keen to do it!

kerem commented

2026-03-03 13:52:46 +03:00

Author

Owner

@datoslabs commented on GitHub (Jun 17, 2025):

Hi,

I recently came across this project and started experimenting with some examples. I've made some enhancements with my local copy to address some issues, more specifically:

Ability to resume after failure/interrupt, at outputDir file level
Added timeout and retry local llm prompting after the process gets stuck on session.promptWithMeta call; when this happens, the process is still running and consuming CPU/GPU resources but the session.promptWithMeta call awaits indefinitely. It appears that the LlmContext needs to be disposed and reinitialized after a few hundred inferences, this allows the process to continue when processing large files.

I think there's a lot of potential for this project. I would be happy to open a PR or create a simple fork as @0xdevalias suggested to house these changes while maintaining the ability to merge/rebase with this project in the future when @jehna returns.

@datoslabs commented on GitHub (Jun 17, 2025): Hi, I recently came across this project and started experimenting with some examples. I've made some enhancements with my local copy to address some issues, more specifically: 1. Ability to resume after failure/interrupt, at `outputDir` file level 2. Added timeout and retry local llm prompting after the process gets stuck on `session.promptWithMeta` call; when this happens, the process is still running and consuming CPU/GPU resources but the `session.promptWithMeta` call awaits indefinitely. It appears that the LlmContext needs to be disposed and reinitialized after a few hundred inferences, this allows the process to continue when processing large files. I think there's a lot of potential for this project. I would be happy to open a PR or create a simple fork as @0xdevalias suggested to house these changes while maintaining the ability to merge/rebase with this project in the future when @jehna returns.

kerem commented

2026-03-03 13:52:46 +03:00

Author

Owner

@0xdevalias commented on GitHub (Jun 17, 2025):

I would be happy to open a PR or create a simple fork as @0xdevalias suggested to house these changes while maintaining the ability to merge/rebase with this project in the future

@datoslabs While it's unlikely to get merged any time soon; for visibility it would be cool to make a PR with your changes. Then we can pull them in more easily when we fork/clean things up.

@0xdevalias commented on GitHub (Jun 17, 2025): > I would be happy to open a PR or create a simple fork as @0xdevalias suggested to house these changes while maintaining the ability to merge/rebase with this project in the future @datoslabs While it's unlikely to get merged any time soon; for visibility it would be cool to make a PR with your changes. Then we can pull them in more easily when we fork/clean things up.

kerem commented

2026-03-03 13:52:46 +03:00

Author

Owner

@datoslabs commented on GitHub (Jun 18, 2025):

@0xdevalias I will spend some time to clean up my code after more testing before creating a PR. My main concern with continuing to create PRs here without the ability to merge into main/dev branches is that over time the project becomes unmaintainable and drive away contributors. A simple fork with active maintainers can ensure PRs get merged in a timely manner before new code/fixes are added.

@datoslabs commented on GitHub (Jun 18, 2025): @0xdevalias I will spend some time to clean up my code after more testing before creating a PR. My main concern with continuing to create PRs here without the ability to merge into main/dev branches is that over time the project becomes unmaintainable and drive away contributors. A simple fork with active maintainers can ensure PRs get merged in a timely manner before new code/fixes are added.

kerem commented

2026-03-03 13:52:46 +03:00

Author

Owner

@0xdevalias commented on GitHub (Jun 18, 2025):

A simple fork with active maintainers can ensure PRs get merged in a timely manner before new code/fixes are added.

@datoslabs Sure, I agree; but are you intending on maintaining that fork? If not, and it just ended up being a fork with your changes on it as well; I feel like that would just fragment things more than an unmerged PR here would.

@0xdevalias commented on GitHub (Jun 18, 2025): > A simple fork with active maintainers can ensure PRs get merged in a timely manner before new code/fixes are added. @datoslabs Sure, I agree; but are you intending on maintaining that fork? If not, and it just ended up being a fork with your changes on it as well; I feel like that would just fragment things more than an unmerged PR here would.

kerem commented

2026-03-03 13:52:46 +03:00

Author

Owner

@datoslabs commented on GitHub (Jun 19, 2025):

I will certainly contribute but I don't have the capacity to maintain a fork alone. The main reason I commented here, instead of creating a new post with the changes I have made, was to gauge if more are interested in a joint effort to maintain a fork that allows others to contribute to this project and move it forward.

@datoslabs commented on GitHub (Jun 19, 2025): I will certainly contribute but I don't have the capacity to maintain a fork alone. The main reason I commented here, instead of creating a new post with the changes I have made, was to gauge if more are interested in a joint effort to maintain a fork that allows others to contribute to this project and move it forward.

kerem commented

2026-03-03 13:52:46 +03:00

Author

Owner

@skilbjo commented on GitHub (Jun 19, 2025):

i started a hard fork but it's immature, currently private. i'm travelling now and won't really have stable housing + quiet time until later this year (i'm in europe for the summer, seeing various founder friends), but i'm happy to invite @0xdevalias and @datoslabs to it and we can build it out in private until we think it's ready for public release. i think in particular @0xdevalias would make a great co-maintainer with me.

i will be able to do a bit of work on it while traveling, but can't become the main steward until things settle down for me.
once it's established, i would be happy to be the main maintainer of the project for the foreseeable future (3+ years), as it's highly relevant to the work i'm doing @ xhrdev (a commercial side project of mine)

@skilbjo commented on GitHub (Jun 19, 2025): i started a hard fork but it's immature, currently private. i'm travelling now and won't really have stable housing + quiet time until later this year (i'm in europe for the summer, seeing various founder friends), but i'm happy to invite @0xdevalias and @datoslabs to it and we can build it out in private until we think it's ready for public release. i think in particular @0xdevalias would make a great co-maintainer with me. i will be able to do a bit of work on it while traveling, but can't become the main steward until things settle down for me. once it's established, i would be happy to be the main maintainer of the project for the foreseeable future (3+ years), as it's highly relevant to the work i'm doing @ xhrdev (a commercial side project of mine)

kerem commented

2026-03-03 13:52:47 +03:00

Author

Owner

@0xdevalias commented on GitHub (Jun 20, 2025):

I will certainly contribute but I don't have the capacity to maintain a fork alone.

@datoslabs nods, yeah, that is totally fair.

The main reason I commented here [..] was to gauge if more are interested in a joint effort to maintain a fork that allows others to contribute to this project and move it forward.

@datoslabs That makes sense. You can see a bit of my / @skilbjo 's thoughts on that a bit earlier in this thread, eg. in https://github.com/jehna/humanify/issues/358#issuecomment-2928512131

Further to @skilbjo 's comments in https://github.com/jehna/humanify/issues/358#issuecomment-2988079058 , I am also currently pretty busy, which is why I haven't progressed on forking the project myself yet. I'm hoping that by mid July or so other things should have slowed down / been able to clear some of the noise / backlog of things currently taking up my time, and at that point I would like to be able to bring more focus to this project.

While I don't currently have a commercial side-project that this is directly relevant to; it is a useful component in an ecosystem of tools that I have been interested in improving for many years (since I was a contracted pentester); and is fairly closely related to some side projects / research ideas that I have been building up notes on for quite a while, and am just waiting for the bandwidth to be able to dive deeper into.

Personally I prefer the idea of there being a sort of 'JavaScript Web App Reverse Engineering' 'community' style GitHub organisation, rather than the fork being hosted on any one individual's personal account. I spent a little time in the past trying to brainstorm name ideas for what that could be (that I have notes on somewhere, though not sure where off the top of my head); but I would also be interested in gathering others ideas on what might make sense for that; as I think it could end up being an interesting/useful 'umbrella org' for other useful tools / project ideas / etc in this space of things.

@0xdevalias commented on GitHub (Jun 20, 2025): > I will certainly contribute but I don't have the capacity to maintain a fork alone. @datoslabs _nods_, yeah, that is totally fair. > The main reason I commented here [..] was to gauge if more are interested in a joint effort to maintain a fork that allows others to contribute to this project and move it forward. @datoslabs That makes sense. You can see a bit of my / @skilbjo 's thoughts on that a bit earlier in this thread, eg. in https://github.com/jehna/humanify/issues/358#issuecomment-2928512131 --- Further to @skilbjo 's comments in https://github.com/jehna/humanify/issues/358#issuecomment-2988079058 , I am also currently pretty busy, which is why I haven't progressed on forking the project myself yet. I'm hoping that by mid July or so other things should have slowed down / been able to clear some of the noise / backlog of things currently taking up my time, and at that point I would like to be able to bring more focus to this project. While I don't currently have a commercial side-project that this is directly relevant to; it is a useful component in an ecosystem of tools that I have been interested in improving for many years (since I was a contracted pentester); and is fairly closely related to some side projects / research ideas that I have been building up notes on for quite a while, and am just waiting for the bandwidth to be able to dive deeper into. Personally I prefer the idea of there being a sort of 'JavaScript Web App Reverse Engineering' 'community' style GitHub organisation, rather than the fork being hosted on any one individual's personal account. I spent a little time in the past trying to brainstorm name ideas for what that could be (that I have notes on somewhere, though not sure where off the top of my head); but I would also be interested in gathering others ideas on what might make sense for that; as I think it could end up being an interesting/useful 'umbrella org' for other useful tools / project ideas / etc in this space of things.

kerem commented

2026-03-03 13:52:47 +03:00

Author

Owner

@0xdevalias commented on GitHub (Oct 11, 2025):

FYI: Given @jehna pinned the following issue, I guess that makes it official that this repo has no intention of being further maintained:

https://github.com/jehna/humanify/issues/639

@0xdevalias commented on GitHub (Oct 11, 2025): FYI: Given @jehna pinned the following issue, I guess that makes it official that this repo has no intention of being further maintained: - https://github.com/jehna/humanify/issues/639 - <img width="314" height="49" alt="Image" src="https://github.com/user-attachments/assets/06ee4f07-fd16-41b8-aeff-e26ebb2e775c" />

kerem commented

2026-03-03 13:52:47 +03:00

Author

Owner

@jehna commented on GitHub (Jan 25, 2026):

@skilbjo did you ever got to making a fork of this project? I think I'll be trying to revive this one again, and would be happy to take in any thoughts you've done differently

@jehna commented on GitHub (Jan 25, 2026): @skilbjo did you ever got to making a fork of this project? I think I'll be trying to revive this one again, and would be happy to take in any thoughts you've done differently

kerem commented

2026-03-03 13:52:47 +03:00

Author

Owner

@skilbjo commented on GitHub (Jan 26, 2026):

hey @jehna thanks for your message.

i started working on a fork, but then i realised AI for anti bot is not going to work, for example things like control flow flattening (google "control flow obfuscation" if unfamiliar to the topic).

the way to do deobfuscation is to do AST analysis with partial evaluation in a VM, only then once it's sufficiently deobfuscated, can you use AI to humanify the code.

probably the best open source deobfuscation project is https://github.com/HumanSecurity/restringer , but this is only part of the problem, you really need to come up with your own custom deobfuscations depending on the anti bot script you are trying to reverse engineer.

but like a great start would be a lib that takes in an anti bot script (like datadome) and uses restringer -> feeds that into humanify.

tldr i realised an AI-only approach is not going to work, or seems like it won't work for a while.

@skilbjo commented on GitHub (Jan 26, 2026): hey @jehna thanks for your message. i started working on a fork, but then i realised AI for anti bot is not going to work, for example things like control flow flattening (google "control flow obfuscation" if unfamiliar to the topic). the way to do deobfuscation is to do AST analysis with partial evaluation in a VM, only then once it's sufficiently deobfuscated, can you use AI to humanify the code. probably the best open source deobfuscation project is https://github.com/HumanSecurity/restringer , but this is only part of the problem, you really need to come up with your own custom deobfuscations depending on the anti bot script you are trying to reverse engineer. but like a great start would be a lib that takes in an anti bot script (like datadome) and uses restringer -> feeds that into humanify. tldr i realised an AI-only approach is not going to work, or seems like it won't work for a while.

kerem referenced this issue

2026-03-03 13:53:15 +03:00

[PR #70] [MERGED] Allow setting non-default OpenAI base URLs #140

Rows
Columns

[GH-ISSUE #358] Open to collaborators? #70