Devr.AI

[amber wagon]

@wheat thorn You are now level 7! :confetti_ball:

[flat dagger]

oh okays, yep that's great.

[empty lintel]

Hi Everyone, I am looking to contribute to Devr.AI . Any tips for an oss beginner will be greatly appreciated.

[warm pulsar]

Hey @eDaud Farzand! 👋 That’s awesome to hear! The best way to start is by picking a small “good first issue” from the repo

[fringe flume]

how are you?

have a nice weekend

[flat dagger]

I'm doing good 🙂

[flat dagger]

Thanks man, you too.
Please try to restrain messages in this channel related to the project only.
Thanks!

[waxen creek]

Hi everyone, I'm new to this project. Could someone please help me set up the codebase? Also, I'm a full-stack developer, so I'm looking for good first issues related to web development—frontend, backend, or full-stack. If the project is primarily about mobile or machine learning, it might not be the best fit for me. Thanks!

[unkempt tulip]

Can we improve the landing pageee

[wheat thorn]

Hi @waxen creek , a setup guide video is in the making and will probably be released today, till then refer to this https://github.com/AOSSIE-Org/Devr.AI/blob/main/docs/INSTALL_GUIDE.md

[waxen creek]

okay but can you tell me is this project good for me as i am having experience in full stack development

please let me know

[wheat thorn]

You can contribute to both the frontend and backend, but the majority of the tasks are backend related, involving langgraph, FastAPI, supabase, graphDB, and more. You should first explore the codebase and identify problems that suits your interests and skills. Here is a snippet of the current tech stack so you can align your interests accordingly

try to start by picking 'good first issue', Our maintainer will soon provide more issues. In the meantime, feel free to look for any bugs or features you’d like to work on

[waxen creek]

thanks for guidance

[warm pulsar]

Hey @flat dagger
https://github.com/AOSSIE-Org/Devr.AI/pull/155
i had resolve the issue and make apr please had a look

[flat dagger]

Thanks @warm pulsar !
I'll review it soon.

[wheat thorn]

@flat dagger Can i work on HIL issue for now if its available

[flat dagger]

yep, you work on it.

Just lmk if you face any issues.

I had already attached majority of the resources for the issue in the issue thread itself

[wheat thorn]

sure! will give u updates of my progress

[wheat thorn]

hey @flat dagger just wanted to confirm that did u faced any import errors when running pytest from the root? I was testing HIL agent state and got some import errors

[wheat thorn]

I moved my test file to the backend folder and ran pytest from there and it worked. But when I ran it from the test folder in the root directory it threw an import error

[oak crypt]

@wheat thorn Setup Guide video released?

[wheat thorn]

Yeah just go to installation guide.md and you will see link on top

[oak crypt]

Can anyone help me I’m facing an issue while setting up the project.
Supabase Environment Variables Not Loading
My FastAPI app (main.py) is failing to start due to a pydantic_core.ValidationError saying that supabase_url and supabase_key are missing.
These values should be loaded from a .env file, but for some reason, python-dotenv isn’t loading them into the environment.

Actually

The app should use python-dotenv and pydantic-settings to load SUPABASE_URL and SUPABASE_KEY from the .env file automatically.

So i started setuping the The .env file is inside the backend/ directory.
It contains both SUPABASE_URL and SUPABASE_KEY with valid values (I’ve double-checked their names and contents).

I Tried:

Verified that the .env file exists and the variable names are correct.

Updated backend/app/core/config/settings.py to explicitly pass the path to .env in load_dotenv().

Printed out debug info:

The current working directory (settings.py location).

The resolved .env file path.

The results of os.getenv("SUPABASE_URL") and os.getenv("SUPABASE_KEY") after loading.

Even though the path is correct, both os.getenv() calls return None.
So, load_dotenv() isn’t loading the variables at all.

Because of this, the Supabase credentials aren’t available and FastAPI throws a validation error during startup.

Error starting server: 2 validation errors for Settings supabase_url Field required [type=missing, input_value={}, input_type=dict] For further information visit https://errors.pydantic.dev/2.11/v/missing supabase_key Field required [type=missing, input_value={}, input_type=dict] For further information visit https://errors.pydantic.dev/2.11/v/missing

[wheat thorn]

dont need to update settings.py explicitly must be some wrong API key issue, regenerate the api key and follow the video guide. If possible provide the ss of that error.

[oak crypt]

yep got it

[flat dagger]

probably some problem with keys only

try setting up referring to the video once again please and just lmk here itself if the issue still persists…

[oak crypt]

yes got resolved

[flat dagger]

cool nice

[amber wagon]

@oak crypt You are now level 2! :confetti_ball:

[oak crypt]

but can u tell which version is best

python version

am getting issues in 3.10

[flat dagger]

those would be just warnings..

not as such of a problem tbh

[oak crypt]

yes i can i share the drafted issue in 3.10 ?

[flat dagger]

ahh btw what to be specific?

[oak crypt]

Issue Title: ImportError: cannot import name 'Self' from 'typing' in Python 3.10 Environments

Description:

When attempting to run the Flask application within the backend/app/database/falkor/code-graph-backend directory, an ImportError occurs, preventing the application from starting. The specific error message is:

ImportError: cannot import name 'Self' from 'typing' (C:\Users\jaswa\AppData\Local\Programs\Python\Python310\lib\typing.py)

Reproduction Steps:

1. Ensure the project's virtual environment is activated and configured for Python 3.10.
2. Navigate to backend/app/database/falkor/code-graph-backend.
3. Attempt to run the Flask application (e.g., flask --app api/index.py run --debug --port 5000).

Expected Behavior:

The Flask application should start successfully without any import errors.

Actual Behavior:

The application fails to start, raising an ImportError because typing.Self cannot be found.

Root Cause Analysis:

The ImportError stems from a compatibility issue with Python versions. The typing.Self type hint was introduced in Python 3.11. The project's current virtual environment, however, is running Python 3.10. When the code attempts to import Self from the typing module in this Python 3.10 environment, the name does not exist, leading to the ImportError.

This ensures the codebase can function correctly across Python 3.10 (minimum supported by pyproject.toml) and newer Python environments.

---

[flat dagger]

oh okays got it

ig the version in doc should be updated to 3.11+

mb for this

[oak crypt]

yeah

[flat dagger]

and thanks a lot too

i personally used somewhat 3.12

[oak crypt]

let me try 3.12 now !

[flat dagger]

I’ll check if the repo can be at-least stabilised for version 3.10>=

[vivid island]

Hey I'm interested in working on the enhancing classification prompt issue (#93)
I saw in the github discussion that you're not approving PRs right now so should I wait, or is it okay to start working on it?

Also, regarding the classification logic: should the bot only respond when explicitly
tagged (@devr) or should it also respond to direct project-related questions
even without a tag?

[wheat thorn]

the bot creates a thread when user inputs its query, there is a node named React classifier which automatically classifies the users Intent we have already done that in React_prompt.py. As you asked we dont need to tag the bot to ask him questions just directly ask in the chat and it will ans you accordingly

[amber wagon]

@wheat thorn You are now level 8! :confetti_ball:

[vivid island]

i see, okay thanks

[waxen creek]

I am new to this community , can any body please let me know about this community and some good first issues
and what are the rules followed like i can raise any issue by approval of maintainer or not
and if i want to work on some issue first ask to maintainter to assign it to u

please guide meeee

[waxen creek]

i had setup all the things and i also signed in the app but what this is happenig

issue in my setup or something else???

I am interested in contributing this project please tell me the areas where i can work

pleasee

I have this console in backend terminal i hope my setup is fine and backend is running properly

@vivid island please see

[waxen creek]

@wheat thorn can you please guide me further

[wheat thorn]

Hi @waxen creek , before raising any issue just let the maintainer know about it here and then raise it, It will be easier for him to get a view of what issue is being raised. Also before working in a new issue you need to ask maintainer first if its in progress or not or even the prerequisites of that issue is completed or not.

[wheat thorn]

tbh i am not sure abt the frontend part you need to confirm with someone else tho your backend is working

[waxen creek]

yeh i think my backend seems fine

[waxen creek]

do u know any one who can tell me about frontend part , and why there is less interaction in this community like i had asked question on 5 , is maintainer is busy ??

[amber wagon]

@waxen creek You are now level 16! :confetti_ball:

[waxen creek]

@wheat thorn please tell me who is the maintainer so that i can ping and ask ?

[wheat thorn]

@flat dagger

[wheat thorn]

Idk abt the community but the maintainer might got some work he was active 3-4 days ago

[waxen creek]

@flat dagger sir can u please tell me about this ???

[flat dagger]

Hey @waxen creek

Actually this UI part way too earlier before building up the Devr.AI platform where we were thinking that the agent could also work in a similar way as coderabbit but later on during the GSoC timeline we didn't went this way and preferred the agent to be mostly around managing the community (Not re-building the wheel (coderabbit)) and so yeah that's why this part doesn't work yet.

[amber wagon]

@flat dagger You are now level 21! :confetti_ball:

[flat dagger]

A fix around this could be simply just send the request to the Falkor DB endpoint which will help in indexing the repositories making the owner for some organization index some of his own preferred repos.

[flat dagger]

You can raise an issue and have fix around this. I'll assign it to you.

My bad, I got involved with a bit of work and yeah will review all the PRs and issues by EOD.
Thank you for your patience!

[wheat thorn]

Hi @flat dagger i was working on HIL integration and a working skeleton is ready like it is responding to my queries correctly tho it doesnt know abt codebase yet. Just wanted to confirm we are using falkorDB for the agent to answer complex queries related to codebase right?

could you provide some docs or any imp info related to falkorDB if needed in this integration

like some specific endpoints

[waxen creek]

@flat dagger sir i am not getting what u want to say ,is your mean is to correct this like in the video show it is not working as i don't know about falkor db , i have to read this
so can u please tell me what the exact issue as i am new to this so it will took some time (not too much) to understand all things

[waxen creek]

like in this as we can see that analyze repo is not working

@flat dagger is making it correct is the task

please let me knoe sir so that i can start work in this

[waxen creek]

@flat dagger please tell me sir so that i can start worl

[vivid island]

@waxen creek follow the setup guide video

[waxen creek]

@waxen creek I had setup properly , i am asking for issue to work on them , i didn't get properly what sir want to say here

[waxen creek]

@flat dagger sir ???

[flat dagger]

it's an endpoint in the current codebase itself.

[flat dagger]

yes, we are using Falkor DB.

And yeah btw it requires repositories to be indexed in order to answer question.

[flat dagger]

Which as of now is via a discord command /index_repository and am preferring to extend this to the UI via this. (Continual to my prev msg)

[flat dagger]

ThankYou for your patience and my bad for late replies. I'll raise a well documented issue.

[flat dagger]

umm, I don't think you need to learn anything in depth about Falkor DB to work with the current system. Just treat it as a MCP server. And what I thought on was HIL to be a general system, preferably can answer nearly every query of user like GPT/LLM does having context and that's all.

[wheat thorn]

ah okays, As of now it is answering basic questions but needs information about the repo. So i just need to index the whole repo (used via falkorDB)

[flat dagger]

Yes

do keep in mind falkor db is limited to a few programming languages as of now (i meant for the codebase indexing facility).

[wheat thorn]

hmm, Got it

[autumn crest]

i have applied for the issue related to enhancing the ui till when will i get the response that should i work on the issue or not

[waxen creek]

i have to learn it , new to falko db

[waxen creek]

sir please open as i want to work as fast i can
and sir one thing is contributing in this project helps regarding gsoc

[waxen creek]

@flat dagger sir after working on one issue i will look issues from myself , but for begining , i need one

[flat dagger]

Stress out bro, I'll document on this tonight.

[flat dagger]

tbh you don't need to have much work-around falkor DB to work on this part. Falkor DB is just a part of our codebase.

If you really wanna explore and learn around is best is to refer their docs and specifically codegraph docs which we are using. They have lot of articles too which are really great, check them out!

[waxen creek]

cool sir

[vivid island]

i'll start working on the bot issue from tomorrow been busy with exams

[amber wagon]

@vivid island You are now level 2! :confetti_ball:

[waxen creek]

ok sir

[flat dagger]

I have raised a well documented issue @waxen creek, you can check it out and just lmk if you face any problems.

I don't think there's any need to learn around FalkorDB to work on the issue. Everything's well implemented, you just need to treat it as an endpoint. That's all.

[wheat thorn]

@flat dagger can u also assign me the HIL issue i have already commented on it

[flat dagger]

oh okays mb

[waxen creek]

Okay sir 🥹

[waxen creek]

sir please assign me too i am working on ti

[waxen creek]

@flat dagger sir one thing i want to ask what to pass in this fucntion as it is asking for a discord id

??

sir for just checking i make it like
async def index_repo(self, repo_input: str, discord_id: str = "web_user") -> Dict[str, Any]:
then i got the dashboard but having wrong data

like in this image u can see that it showing me that 0 pr
but i have around 9 10 pr in devr.ai

@flat dagger let me know sir ?

[waxen creek]

and sir also assign this issue to me
https://github.com/AOSSIE-Org/Devr.AI/issues/171

[flat dagger]

yeah ig cuz it's stored in supabase in UUID format.

can you change it to maybe an optional field?

[flat dagger]

this isn't synced yet tbh

[waxen creek]

Ok

Ok sir i will make it optional

[flat dagger]

currently the majority of work is done around the agent and backend part where we preferred on the agent as of now

[waxen creek]

Just wait a min and I will make pr

[flat dagger]

frontend kind-of just connects the backend and the main thing over here (frontend) is just the bot/agent introduction part by the orgs owners to their respected orgs

[waxen creek]

Cool sir

[flat dagger]

Rest part on frontend will probably be using a separate backend logic if this all is required to be fetched in the frontend too.

[waxen creek]

Sir also assign this to me

[flat dagger]

Take your time. Focus on one issue at a time.

[flat dagger]

I'll review this.

[waxen creek]

Sir that I had done yesterday

Ike there's no much work at that

[flat dagger]

cool

[waxen creek]

@flat dagger sir i had made pr
https://github.com/AOSSIE-Org/Devr.AI/pull/172
please review when u r free

[winged laurel]

Hi 👋
I’m new to open-source and really interested in contributing to Devr.AI. I’ve gone through the repository and would love to get involved could you please suggest some good first issues or small enhancements that would be a good starting point?

Also, I wanted to ask if Devr.AI is planning to participate in Google Summer of Code 2026 (GSoC 2026), or if there’s any plan for it. If yes, I’d love to know about the next set of features or areas you’re focusing on so I can start learning and aligning my contributions in that direction. @left oak @flat dagger

[flat dagger]

sure

[wheat thorn]

hey @flat dagger just did a draft PR, mb i was not able to give it alot of time was busy with my college assignments. Will finish it soon.

[flat dagger]

sure, I'll check it and will notify if any changes are required. Thanks a lot!
Take your time...

[craggy pier]

Hi @flat dagger,
I’m having a few issues getting the backend running smoothly and wanted to double-check if I’m missing something.

The backend container shows as healthy and the basic routes respond fine:

• The user is properly logged in, and I can see the JWT token in the session, but whenever I hit the /integration endpoint, I keep getting authorization failed. I even tried a manual curl request using the same JWT token, but the response is still the same.
• I’ve confirmed the Supabase environment keys, and since the frontend succeeds with the same setup, I’m not sure what’s going wrong on the backend side.
• The /api/repo-stats endpoint returns "detail not found", and I couldn’t find any implementation for this route in router.py or the FastAPI docs within the project.

[flat dagger]

Thanks for notifying @craggy pier, just realized the endpoint path is wrong. The integration routes are mounted at /v1/integrations, not /integration and for the repo-stats endpoint it just doesn't exist in the codebase itself. For the stats of repo the agent itself can answer the queries by the user.

[craggy pier]

If the /api/repo-stats endpoint no longer exists, what is the correct endpoint the frontend should be requesting now? The frontend is currently completely broken because it still depends on that route. and if this is a valid issue could I raise a issue and make a PR for it

[flat dagger]

actually this is probably fixed in https://github.com/AOSSIE-Org/Devr.AI/pull/172/files

which i'm yet to review; I asked this to be changed to use Falkor DB for indexing repos.

[craggy pier]

Ahh ig I'm jobless again

[flat dagger]

Thanks a lot for your consideration. 🙂

[flat dagger]

xD there will not be that much of issues until I refactor the whole codebase once. Need to simplify all of the workflows and preferably try to improve the design of the system.

I'm really sorry, was a bit involved with a few other jobs and my exams too are approaching.

I'll review all PRs and issues when free soon.

[craggy pier]

Np thanks a lot

[raw bone]

hi @flat dagger ,
myself akhil , I am new to the open sources . I would like to work up on Devr.AI . shell we discuss about the new features to be added to it and bugs to be fixed if you are free . are there any meetings going on about the Devr.AI (for development and for new features to be added ) .
if you can suggest some issues which i can work upon.

[keen hornet]

Hi Team,

I'm Eshan Manoj and im setting up the local development environment for the [AOSSIE Devr.AI] repository (frontend).

I ran npm install and npm run dev, but the application is failing with a configuration error because the Supabase keys are missing.

The console error states: Uncaught Error: Missing Supabase configuration. Please set VITE_SUPABASE_URL and VITE_SUPABASE_KEY environment variables.

Could you please provide the development values for these two variables so I can set up my .env file?

VITE_SUPABASE_URL="<value needed>"
VITE_SUPABASE_KEY="<value needed>"
Thanks! I'm ready to start coding once I have these.

[wheat thorn]

Hi eshan, please follow the video guide for setup. You need to manually add your supabase credentials.

[amber wagon]

@wheat thorn You are now level 9! :confetti_ball:

[wheat thorn]

https://github.com/AOSSIE-Org/Devr.AI/blob/main/docs/INSTALL_GUIDE.md

follow this guide

[keen hornet]

alr thnx!!

[wheat thorn]

Hi @flat dagger I encountered an issue in HIL implementation where i need to index a particular repo lets say Devr.AI. In the falkor setup at backend\app\database\falkor\code-graph-backend\api\analyzers\python\analyzer.py the add_dependencies function automatically attempts to pip install all dependencies for every repository it analyzes. In our local environment, this consumes significant space. also the repo_service.py which have the index logic have this clone logic with the /analyze_repo endpoint to falkor. I am thinking instead of cloning it should just read the path as there is already a function /analyze_folder which just POST the file. BUT the analyze_folder only works for local files not github URL only analyze_repo accepts url. Can you tell me should i use analyze repo endpoint only or analyze_folder for development

[flat dagger]

Yep this cloning part was an integral behavior of codegraph itself. We should be using analyze_repo itself only. And yeah, just got an idea, maybe after the repo gets indexed we can schedule a cleanup? Will think a bit more on this and raise an issue.

[wheat thorn]

yeah I was thinking that only cuz everytime you run analyze_repo it clones the whole repo. Would be better if we just delete after the convo ends

[flat dagger]

i think the function is aligned in a way that it won't be re-doing the whole process.

supabase stores metadata for graph indexing

so it'll just notify that this repo has already been indexed

[wheat thorn]

ahhh got it

then i will just continue with analyze_repo

[flat dagger]

cool

[craggy pier]

Hi @flat dagger ,
I created an issue regarding the Profile page that’s currently incomplete on the frontend:Link
If this aligns with the project’s current priorities, could I proceed with working on it?

[wheat thorn]

Hi @flat dagger just asking that have you tried to run analyze_repo before, how long does it took to index. Right now it is taking alot of time, can you confirm me like how long should it take

[flat dagger]

seems good to me. You can work on this.

[flat dagger]

depends on repo size tbh

it took more than 10 mins to index Devr

but now it contains codegraph too and it's size been increase by a lot

you can try on some other small repos too btw

[wheat thorn]

okay

[waxen creek]

@flat dagger hi sir i am back , actually i was busy with my end sem

i will start working on this issue https://github.com/AOSSIE-Org/Devr.AI/issues/171

@flat dagger sir if u have time please review this pr as i had made 2 weeks ago
https://github.com/AOSSIE-Org/Devr.AI/pull/172

[ocean pier]

Hello Everyone
I am Akshaj Tiwari a current CS undergrad.
I am really interested in this project and would love to work in it
Kindly let me know if there is anything necessary which I can contribute to.
Thank you

[acoustic goblet]

read README file and start by setting up project locally

[ocean pier]

Sure

[clever fulcrum]

@acoustic goblet @flat dagger
opeened a issue please assing me
https://github.com/AOSSIE-Org/Devr.AI/issues/176
Status: Open.
#176 In AOSSIE-Org/Devr.AI;· Saurabhyadav0 opened

[stoic wind]

Hi @flat dagger sir,

I found a critical bug in the authentication flow while testing the backend. I've documented the issue and already created a PR with the fix.

Issue #: https://github.com/AOSSIE-Org/Devr.AI/issues/177
**PR #:**https://github.com/AOSSIE-Org/Devr.AI/pull/178#issue-3678444796

Please review when you have time. Thanks!

[flat dagger]

Hey everyone, Thanks a lot for all your contributions. I'm really sorry, I got a bit busy. Will review all the PRs and issues soon.
Thankyou for your patience.

[waxen creek]

@flat dagger sir please review this one

https://github.com/AOSSIE-Org/Devr.AI/pull/172

[waxen creek]

@flat dagger sir i had completed issue 171 and made
https://github.com/AOSSIE-Org/Devr.AI/pull/180
please review sir

[mild surge]

Hi @waxen creek,
I was going through the codebase and encountered the same initialization issue. While reviewing your PR, I noticed a few debug print() statements that seem to have been left in.

[waxen creek]

can you please elaborate more ?

[craggy pier]

@flat dagger Thoughts on moving auth to the backend?
just because it is the standard and we already are connected to the db in the backend.

[mild surge]

Hey @craggy pier are the tests written under tests/ passing successfully on urs or is it just on mine . I am facing issue running the tests.

[craggy pier]

It's better to ask someone else just because my laptop is a toast😅 and given to service.

[mild surge]

also the features discussed on the Readme.md I couldnt find it on the frontend

[dawn acorn]

Hi everyone! I'm Sharanya and I'm completely new to contributing.
I want to start working on Devr.AI, but I’m not sure where to begin.
Can someone help me with the project setup and suggest a simple issue to start with?

[robust birch]

Hi @left oak ! I am highly interested in contributing to Devr.AI for** GSoC 2026**. I have successfully set up the project locally. Is this project likely to be part of GSoC 2026, or should I explore other repo's in AOSSIE?

Btw, regarding the local setup: I faced some #PGRST204 errors with Supabase because tables like interactions and indexed_repositories were missing in the default schema. I verified the database and fixed the SQL script locally to get the backend running perfectly.

@flat dagger If this is a known issue, should I raise a PR with the updated schema to help other contributors set it up faster?"*

[robust birch]

Update: I have raised a PR to fix the missing schema tables amd columns (like indexed_repositories, interactions, etc.) that were causing setup errors.

This should resolve the PGRST204 issues for anyone setting up the project locally.

PR Link: https://github.com/AOSSIE-Org/Devr.AI/pull/181

Thanks!

[mild surge]

Did you run an anlalysis on a repo?

@robust birch

[cyan pilot]

Hey everyone! I’m Vidit, a final-year CSE student. I recently came across this OSS project and found it really interesting, so I’m excited to start contributing. Looking forward to learning and collaborating with you all!

[mild surge]

@flat dagger Sir I have raised a PR issue to update the Install_guide.md file that which were missing steps to setup the Supabase tables which was missing from the Docs and also the flask app was pointed at the wrong location which raised errors . Please review it

https://github.com/AOSSIE-Org/Devr.AI/pull/183#issue-3725781329

@robust birch your PR will not be required as the SQL script already exists but under the supabase/scripts folder hence any changes to the database/integration.sql is not required

[amber wagon]

@mild surge You are now level 2! :confetti_ball:

[verbal oak]

Hi everyone!
I’m Sarabdeep Singh Bilkhu (GitHub: @SarabdeepSBilkhu), a student contributor.

I’ve been exploring Devr.AI locally and getting familiar with the codebase.
Looking forward to learning from the community and contributing where I can.

[mild surge]

Hi @flat dagger @left oak I found a Bug on the frontend side where the landing page would disappear when (as shown in the video in PR)switching between the pages uninitialized moreover the UI would dispaly a simple message to analyze the repo . I have raised an issue and also a PR #188 . Kindly review it and let me know if any changes are to be made .

https://github.com/AOSSIE-Org/Devr.AI/pull/188

[modern shale]

Hi @flat dagger 👋

I’m Parv Mittal. I’m interested in Devr.AI, and I’ve worked on a similar open-source project called Gitmesh, it is related to contributor workflows, analysis, developer telemetry and more. (https://github.com/LF-Decentralized-Trust-labs/gitmesh).

While setting up the project, I ran into several installation issues that aren’t currently reflected in the documentation. I believe these can be streamlined to make onboarding much easier for contributors.

Here are the main issues I noticed, I have also fixed the issue #170 (https://github.com/AOSSIE-Org/Devr.AI/issues/170) :

Dependency conflict
The project lists both py-cord and discord.py as dependencies, which causes a dependency conflict. We can use discord.py only.

GraphDB backend setup (Terminal 3)
The path mentioned in the docs for starting the GraphDB backend is incorrect.
The correct setup requires a different set of dependencies located at backend/app/database/falkor/code-graph-backend.

Supabase setup
The frontend requires a Supabase anon key and project url and the steps to add them are not mentioned in the Install_guide also it is required to manually create tables in Supabase. This process should ideally be automated.

I’m happy to work on fixing these issues to streamline installations so that we could focus on other parts of the project. I have successfully setup the project and solved the issue mentioned in the issue #170 already.

Looking forward to your thoughts — thanks!

[modern shale]

HI maintainers,

I was trying to run and fix the Devr.AI repository for a few days and I have finally created a pull request for it: https://github.com/AOSSIE-Org/Devr.AI/pull/194

This Pull request deals with a lot of things and a lot of code has been written. The project is now fully-functional with better documentation.

I successfully started the dashboard, Contributors, Analytics and pull requests pages and now the project can move onto the next stage.

I need you to review this PR so that I can work on more features. Thanks.

[wary aurora]

Hey @flat dagger , I found a potential thread-safety issue in EmbeddingService—the model can load multiple times on concurrent calls, causing high memory or crashes. Details & proposed fix here: https://github.com/AOSSIE-Org/Devr.AI/issues/193
. I’d like to work on a PR.

[flat dagger]

hey everyone, Apologies for a bit of inactivity from my side.
and yeah, thanks a lot for all the PRs and issues raised. I have started reviewing the work and will provide an update soon to each one of you.
Thanks

[warm pulsar]

@flat dagger Hey ,
if there is some tasks for me then let me know

[warm pulsar]

https://github.com/AOSSIE-Org/Devr.AI/pull/196

i had imporve the ui please have a look

[spiral zodiac]

Hi @left oak 👋
I’m Deepak.

I’ve gone through the documentation and explored the repositories.
I wanted to understand where the project currently stands and which areas are the highest priority right now.

I’m particularly interested in contributing to the Phase-1 components (agent coordination, GitHub/DevRel flow, or tooling), so any guidance on what’s most impactful at this stage would be really helpful.

Thanks!

[remote schooner]

Hello @left oak ,
I’m Swarnesh Kunden, a BE graduate with a Python and ML background.
I’m exploring AOSSIE projects for open-source contribution and GSOC, and I’m currently reviewing beginner-friendly issues to get started.
Looking forward to contributing for GSOC 2026
I would appreciate any suggestions for a good first issue for a new contributor?

[tall quiver]

Hey @flat dagger 👋
I saw that the race-condition fix in EmbeddingService is already being worked on.
I’m planning a small PR to add logging/observability only around model initialization and reuse (no logic or concurrency changes).
Just flagging in case of overlap, happy to adjust if needed.

[remote schooner]

Hi @left oak
I opened a PR fixing a Poetry dependency specification issue that was blocking local setup on my machine. Would appreciate a review when convenient.

[verbal oak]

Hey @left oak
while setting up Devr.AI locally, I found that the Flask-based code-graph backend fails to start on Python 3.10 due to broken import paths and a typing incompatibility.
I opened a PR that fixes the imports and restores local startup without changing behavior. Please let me know if this looks good or needs adjustments.
https://github.com/AOSSIE-Org/Devr.AI/pull/190

[dapper hill]

Hey everyone!
I'm Rudra, a student passionate about AI/ML and Python development. I recently discovered AOSSIE through GSoC and I'm really excited about the projects here!
I'm really interested in Devr.AI. Looking forward to contributing and learning from this amazing community!
Would love to start with a beginner-friendly issue if anyone can point me in the right direction. 🙂
Thank you so much!

[warm pulsar]

Hey Rudra! 👋
Welcome to the community great to have you here!

To get started, check the repo for beginner-friendly / good first issues, and feel free to ask maintainers if you need help picking one.

[flat dagger]

Hey Aditya,
Thanks a lot for helping other contributors out.
I have reviewed your PR. A few changes are needed. Please lmk once do those.
Thanks...

[flat dagger]

I have reviewed a few of the PRs and issues. Can you guys please check on the reviews?
Thanks a lot!
Happy Contributing 🙂

[gaunt vault]

Hey everyone! I’m Siddhant. I recently came across AOSSIE through GSoC and have been exploring the projects here. I’m particularly interested in Devr.AI and excited to start contributing. I’ll be raising an issue soon.

[warm pulsar]

i had done the Requested Changes

[prisma ether]

Hey @sm0kxy I Made an Issue on underlyng problem and then linked a Pull request With it . https://github.com/AOSSIE-Org/Devr.AI/issues/204

[summer ore]

Hello everyone 👋
I’m following up on the Docker-related discussion for the GitHub MCP server (reference #135).

A maintainer asked me to clarify the proposed changes, and I’ve shared a detailed plan on GitHub.

If the approach looks good, could I please be assigned to work on it?
I’ll follow the contribution guidelines and keep the changes scoped and well-documented.

Thank you!

[modern shale]

Thanks! I’ve updated my PR with the requested changes and verified everything is working as expected.

Happy to make further changes if needed.

[waxen creek]

i did the changes , you can review sir
And sir can you tell me the part where i can contribute more

[wheat thorn]

Hi, is anyone facing issues with the Gemini api?? Requests seem to be limited and I cant even ask basic questions to bot. It just shows that the quota is exhausted in console. A month ago it was working perfectly but now i am getting this rate limit issue.

Just wanted to confirm that are you guys using free tier or have enabled billing? like google says i am getting free 300$ worth tokens for 90 days so should i enable it or not

[modern shale]

Hi @wheat thorn
Google has recently reduced free tier limits heavily. So now you only get 20 requests per day for gemini-2.5-flash. They even removed gemini-2.5-pro from the free version.

[modern shale]

I will still use the free tier, might rely on groq api instead if I want more RPD.

[wheat thorn]

Ahh got it... tho somehow my limit is somehow 5 not 20 xD. Will try other free models. Thanks for informing

[solar flame]

Hi everyone !
My name is Sanskriti , a Full Stack & Android developer exploring this organization’s projects.

I’m planning to apply for Google Summer of Code 2026, and I wanted to ask if it is confirmed that this organization will be participating as a mentoring org in GSoC 2026 this year.

If the participation is planned or confirmed, I’d love to start contributing early to active project repositories and prepare a proposal accordingly.

It would be great if you could share any updates or the expected timeline for the official confirmation/announcement.

Looking forward to contributing and being part of the community!
GitHub: https://github.com/Sanskriti10247

Thanks a lot for your guidance

[warm pulsar]

Hi @solar flame , welcome to the community! 👋
Great to see your interest and that you’re thinking ahead for GSOC 2026 that’s definitely the right approach.

As of now, participation for 2026 hasn’t been officially confirmed. Typically, mentoring organizations only get confirmed after Google opens applications for orgs, which usually happens early in the year (around Jan–Feb). Until then, most communities can’t make a formal commitment, even if they intend to apply.

[solar lotus]

Hi everyone! 👋
My name is Mithil, and I’m a beginner frontend developer with knowledge of HTML, CSS, and JavaScript. I’m currently learning by building small practice projects locally and exploring this organization’s projects to understand how open-source collaboration works.

I’m planning to apply for Google Summer of Code 2026, and I wanted to ask whether this organization is planning or considering participating as a mentoring organization for GSoC 2026.

If participation is planned or likely, I would love guidance on how I can start contributing as a beginner, which repositories or issues I should focus on, and what skills I should strengthen to become a good GSoC applicant.

I’m eager to learn, contribute, and grow with the community, and I’d really appreciate any advice or resources you could share.

Thank you for your time and guidance

[mint aurora]

Hi there 👋,
I’m Lokesh Somaiya, a B.Tech Electronics & Computer Science student from VESIT, Mumbai. I’ve been actively working across full-stack development (MERN), AI systems, and backend engineering, along with hands-on cybersecurity industry experience.
I am highly interested in contributing to AOSSIE’s open-source projects and I’m also preparing seriously for Google Summer of Code 2026. I’d love to get involved, understand the project expectations, contribute meaningfully, and learn from the community.

It would be really helpful if you could guide me on:
• How I can begin contributing effectively to AOSSIE
• Suggested repositories or project areas suitable for my tech stack
• Best practices and expectations for first-time contributors

I’m genuinely excited to collaborate, learn, and contribute to impactful work. Your guidance would mean a lot 🙏

[flat dagger]

Hey @stoic wind, I have reviewed your PR. Thanks a lot for this. Can you please do a minor change as mentioned in your PR. 🙂

[flat dagger]

This isn't needed. The scripts for supabase setup are present under backend/app/database/supabase

[flat dagger]

Seems reasonable @mild surge,
could you please review the coderabbit suggested changes once?

[flat dagger]

Seems amazing @modern shale,
I'll review it soon. Thanks a lot 🙂

[flat dagger]

For the dependency conflict part, I can't deny that a very few new users still face dependency issues and packages conflict. I'll try on minimizing these soon after an update to the project from my side. I need to rewrite and align better workflows. Rest, Thanks for your suggestions. I'm open to other suggestions too 🙂

[flat dagger]

Thanks @verbal oak, I'll review the system and your PR soon. We all preferably use 3.11, I'll make sure to update and harden this in the installation guide.

[flat dagger]

The time we started the project Groq had even lower rate limits than Gemini that's why we preferred Gemini API key directly. Thanks for notifying about the rate limits. I'll check and update soon what can be done on this.

[flat dagger]

TBH i was preferring to re-align all the workflows and maybe a better architecture using fewer API calls. I'll try working on this side by side.

I'll update on this or probably may release a design document which can lead to creation of new issues for the contributors.

[flat dagger]

Regarding the API limit problem I found this.
https://www.reddit.com/r/GeminiAI/comments/1pg4et5/comment/nsuh9hu/?utm_source=share&utm_medium=web3x&utm_name=web3xcss&utm_term=1&utm_content=share_button
They'll probably increase the limits soon.

[flat dagger]

You guys can read the whole thread. Pretty informative on Google's plan further ahead on API limits. Might be a usecase for a lot of projects.

[acoustic goblet]

thank you for sharing this

[mild surge]

sure i ll make the changes

[modern shale]

How about having the support of multiple apis? Like OpenAPI, Anthropic, Gemini, Groq etc and even local LLMs?

[verbal oak]

Thanks for taking a look.
I ran into this while setting things up on Python 3.10, which is currently listed as the recommended version in the install guide. Happy to adjust the PR or align with any updated guidance if you decide to standardize on 3.11.

[stoic wind]

done sir can you take a look

[flat dagger]

Local LLMs need a machine for deployment which leads to a cap and a problem to use for everyone.
However, we can try using multiple API keys but that’ll just make code more and more complicated.

[tall quiver]

Hey @flat dagger 👋

Quick sync on the Discord dependency issue we discussed on GitHub (DISCORD_BOT_TOKEN required at startup).
Since the consensus seems to be moving toward a minimal local mode with optional integrations, I wanted to check if you’d prefer me to:

• start with a small code change (make Discord init optional),
• a docs-only PR, or
• both together in a single PR.

Happy to keep it minimal and aligned with whatever direction you finalize.

[wary aurora]

Hi @flat dagger ,

I’ve updated PR #200 https://github.com/AOSSIE-Org/Devr.AI/pull/200 based on the review:

• Clarified lifecycle: the class-level _global_model remains shared, while instance shutdown cleans only per-instance resources. Added shutdown_global() for explicitly releasing the shared model.
• Moved the shutdown check inside the global lock to avoid potential race conditions.
• Tokenizer exceptions are now logged instead of being silently ignored.

All changes are in service.py.
Whenever you have time, could you please review?

Thanks!

[raven bane]

Hello everyone,
I’m Muneer Ali, an MCA student (2025–2027) and a Full-Stack & AI developer based in India. I’m excited to start contributing to AOSSIE and collaborate with the community.
Background
Strong experience with JavaScript, TypeScript, Python, Java
Frontend: React, Next.js, Tailwind CSS
Backend: Node.js, Express, REST APIs
Databases: PostgreSQL, MongoDB, Firebase
Familiar with Git, GitHub workflows, CI/CD
Good foundation in Data Structures & Algorithms
Open-source & project experience
50+ merged pull requests across multiple repositories
Built AI-driven platforms including:
DocMagic – an AI document generation platform
SignifyAI – an Indian Sign Language translation system
Regularly contribute, review PRs, and follow clean commit practices
What I want to contribute
I’m particularly interested in:
Web platforms (React / backend-heavy projects)
Developer tooling
AI/ML-assisted educational or productivity tools
Improving documentation, tests, and scalability
I’m currently exploring AOSSIE repositories and issues and would appreciate guidance on:
Beginner-friendly or “good first issue” tasks
Ongoing projects that need contributors
Best practices followed in AOSSIE contributions
Looking forward to learning, contributing consistently, and collaborating with mentors and fellow contributors.
Thank you!
— Muneer Ali
GitHub: https://github.com/Muneerali199
LinkedIn: https://linkedin.com/in/muneerali

Hey can anyone suggest me some friendly issues

I currently set up project locally

[amber wagon]

@raven bane You are now level 4! :confetti_ball:

[warm pulsar]

Hey Muneer! 👋
Welcome ,
For friendly issues, I’d suggest starting with:
Issues tagged good first issue, beginner, or help wanted
Docs-related issues (README fixes, setup instructions, missing details)

[amber wagon]

@warm pulsar You are now level 8! :confetti_ball:

[warm pulsar]

@flat dagger hey ,
https://github.com/AOSSIE-Org/Devr.AI/pull/155
I had done all the changes requested .

[mild surge]

@flat dagger I have made the changes asked by you . Plz review it

[craggy pier]

Hi @flat dagger Sorry for the delay but here is my plan for implementing the profile page in front end
I’m treating email as the primary identifier for a user’s existence, mainly to avoid unsafe automatic merging between users who register via the frontend and users that already exist from Discord.

When a user registers via the frontend (Supabase Auth), the backend will always create a new application user row keyed by email if one does not already exist. Even if a Discord-only user already exists, no merging will happen at this stage, since email ownership alone does not prove Discord ownership. This intentionally allows a single person to temporarily exist as two rows (Discord-only and email-only), which I consider expected and safe.

If we want both discord and email to be linked Identity unification will only occur through an explicit account-linking flow (e.g. Discord verification). When a Discord user verifies ownership of an email, the backend will update the Discord row to attach the verified email. At that point, multiple rows may share the same email, and the backend can safely aggregate them.

For profile viewing, the backend will not return raw rows. Instead, it will:

• Look up users by verified email
• Aggregate all rows sharing that email into a single profile response
  and the rows without an email remain unmerged until verification occurs.

There are some fields i noticed in the profile page (website, company, Twitter), I plan to add these fields too to the user table.
I think this approach avoids unsafe auto-merging, keeps identity boundaries explicit, and remains backward-compatible with existing Discord-only users.and hopefully not cause any data inconsistencies.

if this sounds good to you i will start working on this issue and submit a PR with 1-3 days.

[raven bane]

Ok thank you very much

[wintry jewel]

Hi everyone 👋
I’m Sainava, a CS undergrad getting familiar with the project.
I’ve done a few open-source contributions before (mainly during Hacktoberfest).
While going through the codebase, I noticed a few incomplete implementations and opened a small PR to replace some empty pass statements and make things more robust.
Happy to be here and looking forward to learning and contributing more!

https://github.com/AOSSIE-Org/Devr.AI/pull/214

[minor siren]

Hey folks,
I’ve opened a small PR improving typing and documentation for the repo URL parsing utility in Devr.AI.
Link: https://github.com/AOSSIE-Org/Devr.AI/pull/215

Would appreciate a review whenever convenient. Thanks!

[signal basin]

Hey everyone 👋

I’m Yash, a beginner in open-source contributions and excited to get started with AOSSIE.

I’m currently learning Python and have basic experience with NumPy, pandas, and Matplotlib. I’m new to contributing to large codebases, but I’m actively looking to learn and improve.

I’d really appreciate any guidance on beginner-friendly issues, documentation/README improvements, or small tasks where I can start contributing and understand the workflow better.

Looking forward to learning from the community and contributing wherever I can 🙂

[lost shadow]

Hey @flat dagger !

I’ve opened a small PR that improves the profile edit UX (clearer input fields and a single save action during edit mode).
Would love your feedback when you get a chance. Thanks!

PR: https://github.com/AOSSIE-Org/Devr.AI/pull/217

[remote schooner]

Hi @left oak ,
I’ve opened a small PR addressing a setup issue I encountered while configuring the project locally as a first-time contributor. The change is scoped to improve the initial setup experience and reduce friction for new contributors.

PR: https://github.com/AOSSIE-Org/Devr.AI/pull/198

I’d really appreciate any feedback when you have time, and I’m happy to make adjustments if needed.
Thanks!

[amber wagon]

@remote schooner You are now level 2! :confetti_ball:

[remote schooner]

hi @flat dagger ,
I attempted to fully run the backend locally and discovered that the codebase currently mixes Weaviate client APIs from different major versions (v3 and v4). This results in runtime attribute/import errors depending on the installed client version. I wanted to surface this before proceeding further, as resolving it likely requires a design decision on which Weaviate version to standardize on.
https://github.com/AOSSIE-Org/Devr.AI/issues/218

[warm pulsar]

Hey i had added the comment please see that

[warm pulsar]

Hey ,
I’ve gone through the changes and left a suggestion please look that

[tall quiver]

Hey, I’ve been going through the Devr.AI design doc and the current backend code, and I want to confirm I understand the intended GitHub integration correctly.

From the guide, it seems GitHub is integrated at two levels:

1. User identity linking via Supabase OAuth (triggered during onboarding /verify_github)
2. System-level capability via a GitHub Agent / GitHub App that the DevRel agent calls as a tool

My question is:

1. Is the expectation that /verify_github only links identity and should not create user profiles before OAuth?
2. And is the actual GitHub “power” (reading repos, commenting, recommendations) expected to come from a GitHub App configured at deployment time, rather than user OAuth tokens?

I just want to make sure I’m aligning my implementation with the intended architecture before going further.

[remote schooner]

Hi @flat dagger , @left oak ,
I’ve opened a second PR that improves the backend setup documentation based on my local setup experience. It adds clearer steps, required dependencies, and notes around current backend startup behavior.

Would appreciate any feedback when you have time — happy to iterate if needed.
Also please let me know if there are any changes required.
https://github.com/AOSSIE-Org/Devr.AI/pull/219

[wintry jewel]

@warm pulsar Hi , thanks for the review . I've made the changes you suggested . Please see if it is okay now .

https://github.com/AOSSIE-Org/Devr.AI/pull/214

[wheat thorn]

Hi guys those who are having problem with gemini api rate limit try to switch to OpenRouter api, its free and i have also tried it. The project is working fine with it. @flat dagger what do u think should we switch to openrouter api or just use it for testing only until gemini increases its limit or we wait for some other solution.

[flat dagger]

yep,
Thanks a lot for an alternative.

It’s better to switch to a different provider.

[ocean oriole]

Hii, can anybody confirm me whether the project setup is working fine with Python 3.10?

[wheat thorn]

it is stable with 3.11

[ocean oriole]

oh thankyou. As the project setup recommended 3.10, so i tried to use 3.10 only but it gave a small error.

[wheat thorn]

install the dependencies using poetry lock file as they have the stable version. Dont update any separately it might causes crashes

[ocean oriole]

got it.

[raven bane]

Hello sir @flat dagger I have created an issue #222 regarding incorrect user profile data after email signup and the need for an editable profile page (excluding email).
Could you please review the issue and let me know if it looks correct and aligned with the project goals ? If approved .l ' d be grateful if you could assign this issues to me so I can start working on it .

[ocean oriole]

Hii @flat dagger @left oak I was testing the Discord → GitHub verification flow locally and noticed an issue where verification can get stuck in a “Verification Pending” state even after the session expires. I’ve opened a GitHub issue for this.
If possible, could you please assign this issue to me? I’d like to work on fixing it.

[remote schooner]

Hi Team, @left oak @flat dagger
I was able to set up the project locally and run the backend until runtime initialization. Execution currently fails due to a Weaviate client API mismatch (v3 vs v4), which requires a maintainer decision to resolve.
I’ve opened a PR and would appreciate maintainer approval before proceeding with a fix.
https://github.com/AOSSIE-Org/Devr.AI/pull/219

[flat dagger]

Hey everyone, I'll try reviewing the issues and PRs tonight. Thank you for your patience. 🙂

[flat dagger]

if possible then can you please share with me the steps to cause such issue?

rest, I'll check once in the system if there exists any problem with token cleanup service...

[ocean oriole]

Yeah sure, but this issue can only be reproduced directly by a user which is not yet verified. (that is, the new users)

The basic reproduction steps include :

1. Run /verify_github from Discord. Bot will create a verification session, and the
   verification_token and verification_token_expires_at will be written to Supabase.
2. Do NOT complete GitHub OAuth. Let the session expire. (wait for more than 5 mins)
3. Run /verify_github again from Discord, and then the bot will display verification pending even after the session expired.

something like this.

This was mainly occuring because the expired token was not getting cleaned properly from the supabase and the /verify_github command was not properly checking the token expiry.

I have made a PR related to this. You may check the changes that I made and can try reproducing the issue before merging.

[amber wagon]

@ocean oriole You are now level 2! :confetti_ball:

[flat dagger]

Hmm, Got it @ocean oriole, I'll try once and then review your PR.
Thanks a lot!

[patent wolf]

Hey I have made a PR for the Devr.AI landing page modifying it and making some UI enhancements can anyone have a look to see if this is good ? https://github.com/AOSSIE-Org/Devr.AI/pull/226

[amber wagon]

@patent wolf You are now level 2! :confetti_ball:

[patent wolf]

Hey folks!
I’m excited to start contributing to Devr.ai and wanted to understand what features or enhancements are still on the roadmap. I’d love to brainstorm ideas and begin contributing, so any guidance on where I can add the most value would be awesome!

[flat dagger]

Hey @patent wolf, I'll let you know on any changes required on your PR. There are already a few PRs on the frontend so you might have to rebase on top of them.
Thanks

[patent wolf]

@flat dagger No prob I'm still modifying my PR though once its completed ilyk then you can give it a review

[raven bane]

Sir @flat dagger have you check issue #222 please let me know if this approach is ok

[raven bane]

??

[orchid ruin]

Hi @flat dagger @left oak , I’m getting started with contributing to Devr.AI and completed the local setup recently. I noticed a small issue in the install guide, opened an issue, and have submitted a PR to fix it.
https://github.com/AOSSIE-Org/Devr.AI/pull/231
Would appreciate a review whenever you have time. Thanks!

[warm pulsar]

https://github.com/AOSSIE-Org/Devr.AI/pull/155

please see this pr @flat dagger its been a long time

[warm pulsar]

Hey i had commented something please see that

[orchid ruin]

Thanks for the review! I’ve replied on the PR, this change fixes the incorrect/missing Flask command path in the docs that causes an import error during setup

[warm pulsar]

@orchid ruin Got it

[patent wolf]

Hey! @flat dagger @left oak Can you review this PR please I have made the enhancements accordingly let me know if you want any other changes to be done https://github.com/AOSSIE-Org/Devr.AI/pull/234

[patent wolf]

@flat dagger @left oak I wanted to ask if Devr.AI would be a part of GSoC 26 or if there are any similar projects that I could explore with a Python based Tech Stack

[tall quiver]

Hey @flat dagger 👋
I opened an issue about repository indexing mutating the environment (dependency installs + venv/ creation) even when running in static mode.
It also seems to sometimes leave failed indexes in a stuck state that can’t be deleted.

I’ve summarized the behavior, evidence, and a possible direction here:
https://github.com/AOSSIE-Org/Devr.AI/issues/237

Would appreciate your thoughts on whether this aligns with the intended indexing guarantees. Thanks!

[amber wagon]

@tall quiver You are now level 2! :confetti_ball:

[patent wolf]

https://github.com/AOSSIE-Org/Devr.AI/issues/241 Hey! @left oak @flat dagger I have found a genuine problem that most LLM API users face I would like to work on it can you guys check it out and assign it to me?

[patent wolf]

@flat dagger @left oak Hey! https://github.com/AOSSIE-Org/Devr.AI/pull/243 Here is the PR to the above issue check it out when you get the time

[topaz swift]

hello i have raised a pr that improves dark mode contrast. By standardizing the color palette using a refined Zinc-based system, this PR ensures that all text elements are readable and interactive components provide clear visual feedback.
It also secured 89/100 score in lighthouse accessibilty score. PR link: https://github.com/AOSSIE-Org/Devr.AI/pull/240#issue-3812172490 would love to recieve a feedback! @flat dagger @left oak

[warm pulsar]

Hey @flat dagger , I’d like to suggest that we have one meeting per week.

[amber wagon]

@warm pulsar You are now level 9! :confetti_ball:

[warm pulsar]

soo that the contributor can resolve their doubt and can contribute

in the better way

[minor siren]

Hey @flat dagger
This is a PR that i have raised a few days ago which was majorly about-

1. "Clear differentiation between retryable and permanent indexing failures (including explicit handling for HTTP 429)"
2. "Fixes a pre-existing bug where the response body could be consumed twice when parsing JSON in aiohttp."

The pr was reviewed both by coderabbit and one of your reviewers and is ready to be merged. Also there's this one suggestion by coderabbit which would take a different and dedicated pr to be worked upon (as it exceeds the scope of this pr). So, u may review the changes of this one which would help me work on the suggested issue in a separate pr.
Thanks !!

https://github.com/AOSSIE-Org/Devr.AI/pull/220

[balmy ingot]

Hey @flat dagger this side Dhruvi. So I was going through the codebase a few days back. I then noticed that there is no option for Google auth. I think for this project it will be great enhancement. And since Supabase supports it so it will be a smooth implementation. I would want to implement this feature and raised an issue for that (link given below). Will be happy for your feedback.

https://github.com/AOSSIE-Org/Devr.AI/issues/245

[balmy ingot]

Well this sounds very good. It will be a great help to understand the project more.

[amber wagon]

@balmy ingot You are now level 6! :confetti_ball:

[acoustic goblet]

Agree

[jade nest]

Hi everyone! 👋

I'm Yashwanth, a Full-Stack Developer (MERN/Next.js) and currently a Software Engineering Intern at Yugayatra. I'm really excited to contribute to Devr.AI, especially given its focus on Agentic workflows and LangGraph!

I have set up the project locally and reviewed the open issues. I would like to work on Issue #244 (Invalid EMBEDDING_MAX_BATCH_SIZE triggers ValueError).

I've analyzed the backend/config.py file and can fix the validation logic to handle the batch size limits correctly. I have already dropped a comment on the GitHub issue.

Looking forward to contributing! 🚀

[jade nest]

PR Submitted! 🚀

I've just opened a PR for Issue #244 (Backend Crash Fix).
Link: https://github.com/AOSSIE-Org/Devr.AI/pull/247

I verified the fix locally, and it properly handles invalid batch sizes now.
Ready for review! Thanks.

[jade nest]

Hi everyone!

While waiting for the review on my first PR (#247), I'd like to pick up a frontend task.
I've commented on Issue #235 (Dynamic Loading States & Validation).

I plan to implement the loading spinners and form validation to improve the UX.
Could someone please assign that to me? Thanks!

[craggy pier]

Hey @flat dagger, the User Profile PR is up! :] It covers the frontend UI, backend merging logic, and the database triggers.

Link: https://github.com/AOSSIE-Org/Devr.AI/pull/248

I added a comment in the description regarding a few changes and updates me know if that looks right to you.

Thanks!

[remote schooner]

Hi, just checking in on this PR #219 in case it was missed. Happy to make updates if needed, or I can move on to another task if preferred.
https://github.com/AOSSIE-Org/Devr.AI/pull/219

[eternal crane]

Hello everyone!My name is Somya Bhadada.I am a fourth year undergraduate student at IIT Kharagpur.I am a beginner in the world of open-source contributions.I wanted some guidance regarding how to choose the issues and how to attempt them, also after submitting the changes usually how many days it takes to get reviewed and merged?

[acoustic goblet]

Hey @eternal crane , you can choose an issue from the Issues tab on GitHub that best suits you and matches your interests. You can start working on it, but please ask for the issue to be assigned to you first. Once a mentor is available, they will review your PR.

[eternal crane]

okay, so should I just ask the mentors in this chat only for the issue to be assigned?

[acoustic goblet]

You can check out GitHub for issues you want to work on, and then ask a mentor to assign one to you.

[floral charm]

Hi everyone 👋
I'm Siddheshwar, a Python backend developer currently exploring the DEVR.AI project.

I’ve been working with Python, FastAPI/Django, and APIs, AI/ML and I’m really interested in how DEVR.AI approaches developer productivity using AI.

I’ve started setting up the project locally and going through the codebase to understand the architecture and workflows.
Looking forward to contributing and learning from the community!

[acoustic goblet]

hey @floral charm welcome to our community. You can start by running the project locally. If u find any bug or issue you can ask mentor to assign it to you..

then you can start working on it

[floral charm]

Ok

[patent wolf]

Hey! can anyone one of the maintainers check this PR out its been a while

@flat dagger

[floral charm]

Hi everyone 👋
I’ve successfully set up the DEVR.AI project locally and explored the codebase a bit.

I’m now looking into the issues and planning to start contributing soon.
Looking forward to working on improving the project!

"I noticed the documentation mentions running flask --app api/index.py run --debug --port 5000 for the backend, but I can't find api/index.py in the main project directory.

Looking at the codebase, the main application uses FastAPI (main.py), and there's a separate Flask service buried in backend/app/database/falkor/code-graph-backend/api/index.py that seems to be for code graph analysis.

Is the Flask command needed for basic Devr.AI functionality, or can I run the project with just:

poetry run python main.py (FastAPI backend)
poetry run python start_github_mcp_server.py (MCP server)
"

[wheat thorn]

no, it is needed to run falkorDB only

[floral charm]

Ok, can you explain little bit

[amber wagon]

@floral charm You are now level 2! :confetti_ball:

[wheat thorn]

basically falkorDB will be used with HIL workflow (currently in progress) which is Graph DB used for extracting files information from the codebase itself to get a better response from the bot.

[amber wagon]

@wheat thorn You are now level 10! :confetti_ball:

[patent wolf]

https://github.com/AOSSIE-Org/Devr.AI/pull/243 Hey @flat dagger @static hazel @left oak Its been a while can you review and merge this PR if possible

[balmy ingot]

Hello @acoustic goblet Sir. I wanted to know can we have a weekly meeting for this project?

[patent wolf]

Hey @flat dagger When you get the time do take a look at this

[patent wolf]

https://github.com/AOSSIE-Org/Devr.AI/issues/251 @flat dagger can you assign this issue to me ?

[warm pulsar]

hello @patent wolf
Nice improvement removing the 100ms polling loop is definitely the right direction

[patent wolf]

@warm pulsar Thanks

[amber wagon]

@patent wolf You are now level 5! :confetti_ball:

[patent wolf]

Can you assign me the issue so I can work more on it ?

[warm pulsar]

@flat dagger you can assign to him
looks good to me

[patent wolf]

@flat dagger @warm pulsar If time permits can you also review this https://github.com/AOSSIE-Org/Devr.AI/pull/243

https://github.com/AOSSIE-Org/Devr.AI/issues/227

@warm pulsar This one was also assigned to me by @flat dagger I have completed it but there is a pending review before merging. If possible can you also take a look when you get the time

[patent wolf]

@flat dagger I'll start working on this to make it production ready then I'll ping you once I raise a PR but don't forget to assign me the issue it will help me to keep track of my assigned contributions : )

[rain monolith]

Hi, I’m preparing a GSoC proposal around extending Devr.AI and have already contributed PR #212. I’d appreciate any feedback on whether my proposed direction aligns with the project roadmap.

[balmy ingot]

Can I know whether I should start working on the implementation of Google Auth? I am asking because I want to know whether mentors want this feature or not in this project?

@flat dagger @warm pulsar

[warm pulsar]

Hey someone will be reviewed soon

[warm pulsar]

@flat dagger What is your opinion ?

[patent wolf]

Hey @warm pulsar do you know when @flat dagger would be up to take a meet for GSOC' 26 Proposal ideas new features etc ?

Like what all does he have in mind for Devr.AI

[warm pulsar]

No idea I think he will be quiet busy

[patent wolf]

Or is there something else planned ?

[warm pulsar]

Its depends
Also this is not a platform to discuss this things here

[patent wolf]

@flat dagger @warm pulsar https://github.com/AOSSIE-Org/Devr.AI/pull/257

[amber wagon]

@patent wolf You are now level 7! :confetti_ball:

[viral wave]

https://github.com/AOSSIE-Org/SocialShareButton/issues/13

[amber wagon]

@viral wave You are now level 4! :confetti_ball:

[viral wave]

https://github.com/AOSSIE-Org/SocialShareButton/pull/14

[warm pulsar]

This coversation is for the Devr.Ai project you should discuss that project related things here only

[amber wagon]

@warm pulsar You are now level 10! :confetti_ball:

[rain monolith]

Got it, thanks for the clarification 🙂
I’ll continue contributing to Devr.AI and refining my GSoC proposal with concrete improvements and PRs.
I’ll share progress here and would appreciate feedback whenever possible.

[rain monolith]

I’ve opened PR #260 adding backend unit tests and fixing a HandlerRegistry bug.
Feedback is welcome whenever convenient.

[wheat thorn]

Hi @flat dagger I have been working on the HIL implementation, and here are some prototype responses from the bot. While the bot was indexing Devr, you had created a class named process_git_history that overloaded the indexing process and caused it to crash as it iterated through all the commits. For now, just for testing, I have commented it out and Devr was indexed successfully. I just need to refine the agent prompts for more robust answers. I wanted to keep you informed about the bot’s responses.

[signal python]

Hii @flat dagger @warm pulsar
https://github.com/AOSSIE-Org/Devr.AI/pull/250

https://github.com/AOSSIE-Org/Devr.AI/pull/255

Please review this pr as per your convenience

[vivid island]

hey this has been up for a while i'd appreciate it if you review it once @warm pulsar @flat dagger
https://github.com/AOSSIE-Org/Devr.AI/pull/191

[warm pulsar]

Add a small test for the classification fallback

[rain monolith]

Hi everyone,
I’ve completed PR #260, which adds comprehensive backend unit test coverage and fixes a handler resolution bug.
All review feedback has been addressed, final cleanup pushed, and all 74 tests pass.
Feedback is welcome whenever convenient thanks!

[signal python]

@warm pulsar @flat dagger please review this prs and ready to change if any thing fails

[ivory umbra]

hi @warm pulsar @flat dagger Please have al ook https://github.com/AOSSIE-Org/Devr.AI/issues/267

[timid geyser]

Hello everyone! I am chanchal ,studying computer science,2nd year student,I came across this project,and it aligns with my tech stack and interest,and i'd love to contribute to this project.i'd love to connect with community members and having suggestions!let me know.thank you

[timid geyser]

Hey everyone and @wraith quarry , is devr.ai not considered for gsoc 2026??

[low axle]

I have the same question too

[rain monolith]

Hi everyone
I’ve opened a GitHub Discussion proposing ideas around reducing PR review bottlenecks using structured validation and AI-assisted checks.
Given the recent announcement about maintenance challenges, I’d appreciate any thoughts or feedback when convenient.
Thanks!

[vapid mist]

Hi everyone,
I explored the backend architecture and reviewed the GitHub webhook handler in backend/routes.py. I noticed the payload is processed without explicit schema validation and structured error handling.

I’ve opened an issue (#275) proposing improvements to validation and robustness of the webhook event processing. Planning to start working on it and submit a PR. Please let me know if this aligns with the project direction or if there are any suggestions before I proceed.

[viscid umbra]

Hi everyone!
I’ve raised a couple of issues recently, and the latest one is about making the profile page editable instead of static. I’d really appreciate your feedback on it before I move ahead with a PR.

This project is something I’m very interested in contributing to, and I’d love to work on this if it aligns with your vision.

Issue link: https://github.com/AOSSIE-Org/Devr.AI/issues/280
Thanks a lot! 🙂

[cedar trout]

"Hi everyone! My name is Ankit Dubey, and I am a 3rd-year Artificial Intelligence & Data Science student. I am excited to connect with the AOSSIE community for GSoC 2026!

I have a strong foundation in C/C++ and Python, with practical experience in Deep Learning and Neural Networks. I recently completed a certification in 'Sequences, Time Series, and Prediction' from DeepLearning.AI, which aligns closely with my interest in advanced modeling.

Recent Work & Skills:

Geospatial AI & Forecasting: I architected a hybrid ConvLSTM-U-Net model to predict wildfire progression, where I engineered a preprocessing pipeline for NetCDF raster data.

Predictive Logistics: For my UIDAI project, I processed 1.5M+ rows of demographic data using HistGradientBoosting to optimize demand forecasting at a granular level.

System Programming: I am comfortable with Linux system APIs and low-level development in C.

I am currently exploring the AOSSIE repositories, particularly projects involving AI/ML and Blockchain (like NeuroTrack or Agora). I look forward to discussing how my experience in spatial intelligence and deep learning can contribute to your innovative 2026 goals!

GitHub: https://github.com/ankiibott

[low axle]

Hi! I was setting up the project locally and ran poetry run pytest from the root.
It looks like tests inside backend/app/database/falkor/code-graph-backend are failing due to import and Python version issues (e.g., typing.Self requiring 3.11+).
Just wanted to confirm — are contributors expected to run and fix those Falkor tests as well, or should we focus only on tests related to the part of the codebase we modify?

Thanks!

[warm pulsar]

Hey Ankit! 👋
Welcome to AOSSIE great to have you here!

[cedar trout]

Hi everyone,
I've successfully set up the Devr.AI project locally on my system. It's running great!
However, during the setup process, I noticed a couple of things that might need attention:

Backend Dependencies (requirements.txt): The requirements.txt file specifies crewai==0.120.1, but this specific version seems to be unavailable (at least on standard PyPI repositories), causing pip install to fail. I resolved this locally by relaxing the version constraint to just crewai (or upgrading to a newer version like 0.28.0+).

(Suggestion: We might want to update the requirement to a currently available version.)

Frontend Configuration: I noticed the environment example file in the frontend directory is named .env example (with a space). Standard convention is usually .env.example or .env.local.example. This might be a minor typo but could confuse automated scripts or new contributors.

(Suggestion: Rename frontend/.env example to frontend/.env.example.)

I'd be happy to open a PR to fix these if that helps!

Best, Ankit Dubey
@pliant heath @wraith quarry @warm pulsar

[warm pulsar]

Hi Ankit,
Thank you so much for setting up Devr.AI locally and for the detailed really appreciate you taking the time to point this out!
once the maintainer approve you can start working

[rain monolith]

Hi everyone

My name is Mohammad Tihame and I’m a 3rd-year B.Tech student in Artificial Intelligence. I’m excited to actively contribute to AOSSIE and Devr.AI for GSoC 2026.

I’ve been contributing to Devr.AI recently (PR #260 – backend core test coverage and handler bug fix) and I’m particularly interested in improving reliability, AI integrations, and scalable backend systems.

I have a strong foundation in Python and system-level programming. I’ve built ML systems using LightGBM, XGBoost, and deep learning pipelines. I’ve worked with large synthetic datasets (5M+ rows) for prediction systems and built AI-powered platforms including medical imaging diagnostics, RAG agents, and local AI systems. I’m also comfortable with backend testing, async systems, and architecture improvements.

I’m currently exploring areas where I can improve AI robustness and rate-limit handling, strengthen backend reliability and test coverage, and propose scalable AI-assisted workflows for maintenance.

I’d love to understand which areas need contributors most right now and how I can align my GSoC proposal with long-term project impact.

Looking forward to collaborating 🚀
GitHub: https://github.com/muhammadtihame

[timid geyser]

@warm pulsar are you mentor of devr ai or contributer?

[craggy pier]

He's a contributor the maintainer is @flat dagger

[desert shard]

Hi everyone 👋

I'm Mahek, currently contributing to Devr.AI.

I recently raised and worked on a documentation improvement related to installation setup, Python version compatibility, and dependency clarification to reduce onboarding friction.

I’ve been setting up both frontend and backend locally, exploring Supabase authentication flow, and reviewing UI-related issues to identify beginner-friendly contributions.

Currently focusing on frontend polish and small UX enhancements.

Excited to collaborate and contribute consistently!

[balmy ingot]

Hey @warm pulsar What do you think about this?

https://github.com/AOSSIE-Org/Devr.AI/issues/245

[vivid island]

done, can you please check it out!

[warm pulsar]

it was a good idea

but i think they are already using the supabase
so in that there is an option itself to login with google.

[unkempt shard]

https://github.com/AOSSIE-Org/Devr.AI/issues/284 hi sir @warm pulsar pls review

[amber wagon]

@unkempt shard You are now level 4! :confetti_ball:

[rain monolith]

Hi everyone

Following the recent suggestion about performing human end-to-end reviews, I’ve started a structured audit of Devr.AI from a user perspective to identify completion gaps in existing features (not proposing new features).

I’ve opened one issue related to silent 401 handling and will continue reviewing flows carefully to document meaningful stabilization gaps.

If there’s any preferred format or priority area I should focus on, I’d appreciate guidance.

Thanks.

[desert shard]

Hi 👋

I’ve commented on issue #284 and would like to start implementing the rate limiting with exponential backoff.

Before proceeding, I just want to confirm the preferred approach:

Should this be implemented as a wrapper around the existing Discord client?

Or would you prefer integrating retry logic directly into the current bot layer?

Also, is Redis already configured in staging, or should I implement a local in-memory version first and extend later?

I’m planning to start with a minimal retry + backoff implementation and then extend to distributed tracking if needed.

Looking forward to your guidance 🚀

[amber wagon]

@desert shard You are now level 2! :confetti_ball:

[raven bane]

If anyone is stuck on an issue or needs help reviewing or debugging something, feel free to tag me or assign it. I’m happy to collaborate and support where I can.

[static notch]

Hey everyone, I'm just jumping into the Devr.AI side of AOSSIE for GSoC, really excited to start contributing with you all.

I saw @final flicker's org-wide announcement earlier about standardizing the CodeRabbit AI configuration. I noticed Devr.AI hadn't received the update yet, so I went ahead and dropped the .coderabbit.yaml template file into the repo so our CI/CD stays perfectly aligned with the rest of the org.

Would really appreciate a quick review/merge whenever the maintainers have a moment! 🙏
PR: https://github.com/AOSSIE-Org/Devr.AI/pull/287

[amber wagon]

@static notch You are now level 3! :confetti_ball:

[desert shard]

Hi 👋

I’ve updated PR #288 for Issue #284 and resolved merge conflicts.

CodeRabbit suggested improvements for Markdown formatting and formula consistency in documentation.

Before I push further changes, just wanted to confirm:
Should I update the documentation formatting as suggested, or is the current structure acceptable?

Thanks!
PR Link: https://github.com/AOSSIE-Org/Devr.AI/pull/288

[warm pulsar]

Hii ,
Regarding CodeRabbit’s suggestions: I’d recommend going ahead and applying the improvements for Markdown formatting and formula consistency

[unkempt shard]

@warm pulsar sir this is my issue #issue 284

[warm pulsar]

dont call me sir
i am just an another contributor like you

[amber wagon]

@warm pulsar You are now level 11! :confetti_ball:

[unkempt shard]

It's ok but I made issue and my implementation is ready but someone all ready do pr and say close #284

[amber wagon]

@unkempt shard You are now level 5! :confetti_ball:

[unkempt shard]

I think it is good if we work on our own issue because understanding of our own created issue is much better

[warm pulsar]

we can't stop anyone to working on the issue
you can create your PR and its maintainer choice to choose the Best

[unkempt shard]

Hi sir @final flicker when you are free please see https://github.com/AOSSIE-Org/Devr.AI/pull/289

[desert shard]

Hi 👋

I’ve completed Issue #278 (Footer Visual Hierarchy Improvement) and opened a PR for it:

🔗 https://github.com/AOSSIE-Org/Devr.AI/pull/288

The changes focus on improving horizontal space distribution and layout structure while keeping branding intact.

Would appreciate a review when possible. Thank you!

[waxen summit]

hey , i really liked this project, can u guys help me how i can start contributing ?

[wheat thorn]

Hi Ved, First, fork the project and clone your forked repository to your local machine. Follow the documentation to set up the bot (if you face any issues, feel free to ask here). To contribute you can either fix a bug or implement a new feature. Once done, create a pull request and discuss it with the mentor.

[final flicker]

I do not mentor for or maintain DevR.AI. Please contact <@&1371443573584826489> .

@flat dagger .

[unkempt shard]

Ok sir

@flat dagger
https://github.com/AOSSIE-Org/Devr.AI/pull/289

[rain monolith]

I have submitted the draft proposal and would appreciate any feedback from mentors.

[rain monolith]

Hi everyone ,

Quick update on my PR improving backend error handling and async reliability.

I’ve addressed the CodeRabbit review suggestions, including:

1. safer API error responses (no internal exception leakage)
2. background task tracking in the event bus
3. proper exception propagation in queue workers
4. prevention of UnboundLocalError in the coordinator
5. reliable Discord response handling
6. GitHub webhook signature verification

All modified files compile successfully and the existing tests remain unaffected.

PR: https://github.com/AOSSIE-Org/Devr.AI/pull/292

Happy to make further adjustments if needed. Thanks.

[wary aurora]

Hi @flat dagger , hope you're doing well.

I just wanted to kindly follow up on my PR. When you get time, could you please take a look and share any feedback?

PR: https://github.com/AOSSIE-Org/Devr.AI/pull/200

Thank you!

[inland herald]

hii @flat dagger , i want to work on this issue . feel free to assign me sir
https://github.com/AOSSIE-Org/Devr.AI/issues/294

[raven bane]

Hi sir @flat dagger , could you please take a look at this issue when you have time? I believe it might need attention. If possible, I’d be happy to work on fixing it. Please feel free to assign it to me whenever convenient.

Issue: https://github.com/AOSSIE-Org/Devr.AI/issues/295

Thank you!