From fa7571c415bd1c1eb1de99c4bf9233b83badff8e Mon Sep 17 00:00:00 2001 From: peplxx <91543105+peplxx@users.noreply.github.com> Date: Tue, 10 Jun 2025 12:02:17 +0300 Subject: [PATCH 01/10] docs: init _index.md file --- content/docs/2025/scaffold/_index.md | 4 ++++ 1 file changed, 4 insertions(+) create mode 100644 content/docs/2025/scaffold/_index.md diff --git a/content/docs/2025/scaffold/_index.md b/content/docs/2025/scaffold/_index.md new file mode 100644 index 000000000..38b987b43 --- /dev/null +++ b/content/docs/2025/scaffold/_index.md @@ -0,0 +1,4 @@ +--- +bookCollapseSection: true +title: "Scaffold" +--- \ No newline at end of file From d77546b86c196b5df619e57350b042282a8070f0 Mon Sep 17 00:00:00 2001 From: peplxx <91543105+peplxx@users.noreply.github.com> Date: Tue, 10 Jun 2025 12:06:40 +0300 Subject: [PATCH 02/10] chore: change root dir --- content/docs/2025/{scaffold => beer-bears-scaffold}/_index.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) rename content/docs/2025/{scaffold => beer-bears-scaffold}/_index.md (52%) diff --git a/content/docs/2025/scaffold/_index.md b/content/docs/2025/beer-bears-scaffold/_index.md similarity index 52% rename from content/docs/2025/scaffold/_index.md rename to content/docs/2025/beer-bears-scaffold/_index.md index 38b987b43..18e348d80 100644 --- a/content/docs/2025/scaffold/_index.md +++ b/content/docs/2025/beer-bears-scaffold/_index.md @@ -1,4 +1,4 @@ --- bookCollapseSection: true -title: "Scaffold" +title: "Beer Bears: Scaffold" --- \ No newline at end of file From cdd9dd8265206120b6c45e0d4c3112619173a771 Mon Sep 17 00:00:00 2001 From: peplxx <91543105+peplxx@users.noreply.github.com> Date: Tue, 10 Jun 2025 12:11:27 +0300 Subject: [PATCH 03/10] chore: leave only team name --- content/docs/2025/{beer-bears-scaffold => beer-bears}/_index.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) rename content/docs/2025/{beer-bears-scaffold => beer-bears}/_index.md (52%) diff --git a/content/docs/2025/beer-bears-scaffold/_index.md b/content/docs/2025/beer-bears/_index.md similarity index 52% rename from content/docs/2025/beer-bears-scaffold/_index.md rename to content/docs/2025/beer-bears/_index.md index 18e348d80..f46c580c6 100644 --- a/content/docs/2025/beer-bears-scaffold/_index.md +++ b/content/docs/2025/beer-bears/_index.md @@ -1,4 +1,4 @@ --- bookCollapseSection: true -title: "Beer Bears: Scaffold" +title: "Beer Bears" --- \ No newline at end of file From eca5aca34e61fd27eeef517c711f79819f712c07 Mon Sep 17 00:00:00 2001 From: Dmitry Prosvirkin <115407430+Trunn5@users.noreply.github.com> Date: Tue, 10 Jun 2025 19:41:43 +0300 Subject: [PATCH 04/10] Create week1.md --- content/docs/2025/beer-bears/week1.md | 73 +++++++++++++++++++++++++++ 1 file changed, 73 insertions(+) create mode 100644 content/docs/2025/beer-bears/week1.md diff --git a/content/docs/2025/beer-bears/week1.md b/content/docs/2025/beer-bears/week1.md new file mode 100644 index 000000000..16993bdf7 --- /dev/null +++ b/content/docs/2025/beer-bears/week1.md @@ -0,0 +1,73 @@ +--- +title: "Week #1" +--- + +# Week #1 + +## Project description + +### Project name: Scaffold + +**Code repository**: https://github.com/IU-Capstone-Project-2025/LINK_TO_YOUR_REPOSITORY <- ???? + +Scaffold is a temporary structure used to support a work crew and materials to aid in the construction, maintenance and repair of buildings + +Our Scaffold is a code management system designed to translate and maintain source code as a graph in a graph database, +enabling seamless context injection for large language models (LLMs). And helps AI agents in construction, maintenance and repair of your project. + +### **Team Members** + +| Team Member | Telegram Alias | Email Address | Track | Responsibilities | +| :-------------- | :--------------- | :---------------------- | :------------ | :---------------------------------------- | +| Melnikov Sergei | @peplxx | s.melnikov@innopolis.university | Project Owner | Team Management, RAG Algorithms | +| Razmakhov Serhei | @onemoreslacker | s.razmakhov@innopolis.university | Developer | Languages parsers, AT Generation | +| Prosvirkin Dmitry | @dmitry5567 | d.prosvirkin@innopolis.university | Developer | Vector, Graph Database Management| +| Mashenkov Timofei | @mashfeii | t.mashenkov@innopolis.university | Developer | Context Fethcing Algotihm | +| Glazov Sergei | @pushkin404 | s.glazov@innopolis.university | QA | QA Research, MCP Analysis| + +## Brainstorming + +### Ideas during brainstorming + +*Put here your ranked ideas with short description* + +### Brief market research / problem validation + +*Put here a brief market research/problem validation for 1–2 ideas* + + +## Basic requirements + +### Target users and their primary needs + +*...* + +### User stories + +*...* + +### Initial scope + +*...* + + +## Tech-stack + +*...* + +## *Something else you want to add* + +*Feel free to add anything else that you consider important to your report* + + +# Weekly commitments + +## Individual contribution of each participant + +*...* + +## Confirmation of the code's operability + +We confirm that the code in the main branch: +- [ ] In working condition. +- [ ] Run via docker-compose (or another alternative described in the `README.md`). From 66428ec0d1c50d547fba87169d387f354624c374 Mon Sep 17 00:00:00 2001 From: Dmitry Prosvirkin <115407430+Trunn5@users.noreply.github.com> Date: Tue, 10 Jun 2025 20:18:41 +0300 Subject: [PATCH 05/10] Update week1.md --- content/docs/2025/beer-bears/week1.md | 54 +++++++++++++++++++++++---- 1 file changed, 47 insertions(+), 7 deletions(-) diff --git a/content/docs/2025/beer-bears/week1.md b/content/docs/2025/beer-bears/week1.md index 16993bdf7..c65c75d3c 100644 --- a/content/docs/2025/beer-bears/week1.md +++ b/content/docs/2025/beer-bears/week1.md @@ -38,33 +38,73 @@ enabling seamless context injection for large language models (LLMs). And helps ## Basic requirements +Parse code into AST and build code graphs + +Store in a graph DB (e.g., Neo4j) and vector DB (e.g., Qdrant) + +Extract structural/code entity relationships (calls, imports, etc.) + +Provide API/CLI for context queries + +Support incremental updates (e.g., Git hooks or file watchers) + +Enable context injection into LLMs (RAG) + +Basic testing and validation tools + ### Target users and their primary needs -*...* +Developers Understand and refactor code faster using AI and graph context +AI Engineers Provide structured context to LLMs for better accuracy +Tech Writers Auto-generate or update documentation from code structure +QA Engineers Understand dependencies and test impact of code changes ### User stories -*...* +As a developer, I want to find all references to a function to safely rename it. + +As an AI engineer, I want structured code context to improve RAG results. + +As a tech writer, I want to auto-generate docs from code relationships. + +As a QA engineer, I want to trace service dependencies for better test coverage. ### Initial scope -*...* +Python code parser → graph + vector DB + +Neo4j + Qdrant integration + +Basic API/CLI for context lookup + +LLM context injection (early RAG prototype) + +CLI tool for developers + +Basic graph update system (e.g., file watcher) ## Tech-stack -*...* +Python – Widely used in AI and tooling; ideal for building parsers, integrating LLMs, and rapid prototyping. + +Neo4j – Purpose-built graph database optimized for modeling and querying complex code relationships. -## *Something else you want to add* +VectorDB (e.g., Qdrant) – Enables high-performance semantic search over embedded code/document chunks. -*Feel free to add anything else that you consider important to your report* +Docker – Provides consistent, containerized environments for development, testing, and deployment. +LLM Chain (e.g., LangChain) – Modular framework for orchestrating Retrieval-Augmented Generation pipelines. # Weekly commitments ## Individual contribution of each participant -*...* +Melnikov Sergei - brainstorming, repository, informtaion research +Razmakhov Serhei- brainstorming, repository +Prosvirkin Dmitry - brainstorming, writing report +Mashenkov Timofei - brainstorming, informtaion research +Glazov Sergei - brainstorming, writing report ## Confirmation of the code's operability From c393380df176336885b0d9a721c6cda64a4b9913 Mon Sep 17 00:00:00 2001 From: Dmitry Prosvirkin <115407430+Trunn5@users.noreply.github.com> Date: Tue, 10 Jun 2025 20:29:45 +0300 Subject: [PATCH 06/10] Update week1.md --- content/docs/2025/beer-bears/week1.md | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/content/docs/2025/beer-bears/week1.md b/content/docs/2025/beer-bears/week1.md index c65c75d3c..3cd4d76da 100644 --- a/content/docs/2025/beer-bears/week1.md +++ b/content/docs/2025/beer-bears/week1.md @@ -8,7 +8,9 @@ title: "Week #1" ### Project name: Scaffold -**Code repository**: https://github.com/IU-Capstone-Project-2025/LINK_TO_YOUR_REPOSITORY <- ???? +**Code repository**: + - [Course Repository](https://github.com/IU-Capstone-Project-2025/scaffold) + - [Our Repository](https://github.com/Beer-Bears/scaffold) Scaffold is a temporary structure used to support a work crew and materials to aid in the construction, maintenance and repair of buildings From 3e2d2f8b2fd78d7caf397e814ca74c47e034e9a7 Mon Sep 17 00:00:00 2001 From: Dmitry Prosvirkin <115407430+Trunn5@users.noreply.github.com> Date: Tue, 10 Jun 2025 20:31:34 +0300 Subject: [PATCH 07/10] Update week1.md --- content/docs/2025/beer-bears/week1.md | 22 ++++++++++++++++++++-- 1 file changed, 20 insertions(+), 2 deletions(-) diff --git a/content/docs/2025/beer-bears/week1.md b/content/docs/2025/beer-bears/week1.md index 3cd4d76da..4a0c60201 100644 --- a/content/docs/2025/beer-bears/week1.md +++ b/content/docs/2025/beer-bears/week1.md @@ -31,12 +31,30 @@ enabling seamless context injection for large language models (LLMs). And helps ### Ideas during brainstorming -*Put here your ranked ideas with short description* +1 Graph-based code context platform for LLMs +Translate source code into a graph database (AST, function/class relations) to serve as rich structured context for AI agents. Enables scalable, accurate retrieval of relevant information for code generation and QA. + +2 AI codebase companion (Scaffold CLI) +CLI tool integrated into developer workflows that allows querying, summarizing, or modifying the codebase using LLMs with graph-backed context. + +3 LLM-aware refactoring assistant +Leverages code graphs and embeddings to propose or automate safe refactoring operations (rename symbols, split/merge functions, remove dead code). ### Brief market research / problem validation -*Put here a brief market research/problem validation for 1–2 ideas* +Idea 1: Graph-based code context platform for LLMs +Problem: Modern LLMs operate on tokenized text and lack awareness of the structural and semantic organization of real-world codebases. Existing solutions (e.g., embedding chunks into a vector DB) do not capture hierarchical or reference-based relationships well. + +Existing solutions: Tools like Sourcegraph Cody, Codeium, and GitHub Copilot use text embeddings but struggle with large-scale project structure and maintaining long-term context. + +Validation: Research from OpenAI, Meta, and others highlights the importance of hierarchical and symbolic context in improving AI performance on large-scale code reasoning tasks. Graph-based representations are also used in tools like CodeQL for similar reasons. + +Idea 2: LLM-aware refactoring assistant +Problem: Refactoring at scale (e.g., renaming a core service method used in hundreds of files) is high-risk and hard to reason about, especially across language boundaries. + +Existing solutions: IDEs like IntelliJ or VSCode offer local static analysis refactors, but not AI-assisted reasoning or graph-level semantic refactoring. +Validation: Enterprise engineering teams report significant friction in large-scale refactoring, especially when team members are unfamiliar with legacy code or there’s poor documentation. GitHub Copilot lacks this structured reasoning. ## Basic requirements From 472929e23c90e0c3685fd9e26e0adbbc398910ee Mon Sep 17 00:00:00 2001 From: Dmitry Prosvirkin <115407430+Trunn5@users.noreply.github.com> Date: Wed, 11 Jun 2025 17:36:14 +0300 Subject: [PATCH 08/10] Update week1.md --- content/docs/2025/beer-bears/week1.md | 24 ++++++++++++++++++------ 1 file changed, 18 insertions(+), 6 deletions(-) diff --git a/content/docs/2025/beer-bears/week1.md b/content/docs/2025/beer-bears/week1.md index 4a0c60201..124b472dc 100644 --- a/content/docs/2025/beer-bears/week1.md +++ b/content/docs/2025/beer-bears/week1.md @@ -23,7 +23,7 @@ enabling seamless context injection for large language models (LLMs). And helps | :-------------- | :--------------- | :---------------------- | :------------ | :---------------------------------------- | | Melnikov Sergei | @peplxx | s.melnikov@innopolis.university | Project Owner | Team Management, RAG Algorithms | | Razmakhov Serhei | @onemoreslacker | s.razmakhov@innopolis.university | Developer | Languages parsers, AT Generation | -| Prosvirkin Dmitry | @dmitry5567 | d.prosvirkin@innopolis.university | Developer | Vector, Graph Database Management| +| Prosvirkin Dmitry | @dmitry5567 | d.prosvirkin@innopolis.university | Developer | Vector, Graph Database Management | | Mashenkov Timofei | @mashfeii | t.mashenkov@innopolis.university | Developer | Context Fethcing Algotihm | | Glazov Sergei | @pushkin404 | s.glazov@innopolis.university | QA | QA Research, MCP Analysis| @@ -120,11 +120,23 @@ LLM Chain (e.g., LangChain) – Modular framework for orchestrating Retrieval-Au ## Individual contribution of each participant -Melnikov Sergei - brainstorming, repository, informtaion research -Razmakhov Serhei- brainstorming, repository -Prosvirkin Dmitry - brainstorming, writing report -Mashenkov Timofei - brainstorming, informtaion research -Glazov Sergei - brainstorming, writing report +Melnikov Sergei + - brainstorming, + - [READme and structure of repository](https://github.com/Beer-Bears/scaffold/pull/2), + - informtaion research +Razmakhov Serhei + - brainstorming, + - repository +Prosvirkin Dmitry + - brainstorming, + - writing report + - [Google Docs description](https://docs.google.com/document/d/1K4CPKvia2kNnlKm9MNFnxmQRqHM1KS_lJMJzueEnQVE/edit?usp=sharing) +Mashenkov Timofei + - brainstorming, + - informtaion research +Glazov Sergei + - brainstorming, + - writing report ## Confirmation of the code's operability From a62d5f0303d40386d637fa5b440203a2ea066153 Mon Sep 17 00:00:00 2001 From: Dmitry Prosvirkin <115407430+Trunn5@users.noreply.github.com> Date: Wed, 11 Jun 2025 17:38:01 +0300 Subject: [PATCH 09/10] Update week1.md --- content/docs/2025/beer-bears/week1.md | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-) diff --git a/content/docs/2025/beer-bears/week1.md b/content/docs/2025/beer-bears/week1.md index 124b472dc..eb8cfeed3 100644 --- a/content/docs/2025/beer-bears/week1.md +++ b/content/docs/2025/beer-bears/week1.md @@ -123,20 +123,25 @@ LLM Chain (e.g., LangChain) – Modular framework for orchestrating Retrieval-Au Melnikov Sergei - brainstorming, - [READme and structure of repository](https://github.com/Beer-Bears/scaffold/pull/2), + - [Excalidraw Scheme](https://excalidraw.com/#json=8DxFWGT66eisYHShw5etd,FTPXATItDi33jkpVtVr54A), - informtaion research -Razmakhov Serhei +Razmakhov Sergey - brainstorming, + - [Excalidraw Scheme](https://excalidraw.com/#json=8DxFWGT66eisYHShw5etd,FTPXATItDi33jkpVtVr54A), - repository Prosvirkin Dmitry - brainstorming, - - writing report + - writing report, + - [Excalidraw Scheme](https://excalidraw.com/#json=8DxFWGT66eisYHShw5etd,FTPXATItDi33jkpVtVr54A), - [Google Docs description](https://docs.google.com/document/d/1K4CPKvia2kNnlKm9MNFnxmQRqHM1KS_lJMJzueEnQVE/edit?usp=sharing) Mashenkov Timofei - brainstorming, - - informtaion research + - [Excalidraw Scheme](https://excalidraw.com/#json=8DxFWGT66eisYHShw5etd,FTPXATItDi33jkpVtVr54A), Glazov Sergei - brainstorming, - - writing report + - informtaion research, + - [Excalidraw Scheme](https://excalidraw.com/#json=8DxFWGT66eisYHShw5etd,FTPXATItDi33jkpVtVr54A), + ## Confirmation of the code's operability From 87e99058891b820dff6f6209c4b9b87aae3d6a75 Mon Sep 17 00:00:00 2001 From: Dmitry Prosvirkin <115407430+Trunn5@users.noreply.github.com> Date: Wed, 11 Jun 2025 18:56:55 +0300 Subject: [PATCH 10/10] Update week1.md --- content/docs/2025/beer-bears/week1.md | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/content/docs/2025/beer-bears/week1.md b/content/docs/2025/beer-bears/week1.md index eb8cfeed3..2bf0cb4f4 100644 --- a/content/docs/2025/beer-bears/week1.md +++ b/content/docs/2025/beer-bears/week1.md @@ -125,18 +125,22 @@ Melnikov Sergei - [READme and structure of repository](https://github.com/Beer-Bears/scaffold/pull/2), - [Excalidraw Scheme](https://excalidraw.com/#json=8DxFWGT66eisYHShw5etd,FTPXATItDi33jkpVtVr54A), - informtaion research + - Razmakhov Sergey - brainstorming, - [Excalidraw Scheme](https://excalidraw.com/#json=8DxFWGT66eisYHShw5etd,FTPXATItDi33jkpVtVr54A), - repository + - Prosvirkin Dmitry - brainstorming, - writing report, - [Excalidraw Scheme](https://excalidraw.com/#json=8DxFWGT66eisYHShw5etd,FTPXATItDi33jkpVtVr54A), - [Google Docs description](https://docs.google.com/document/d/1K4CPKvia2kNnlKm9MNFnxmQRqHM1KS_lJMJzueEnQVE/edit?usp=sharing) + Mashenkov Timofei - brainstorming, - [Excalidraw Scheme](https://excalidraw.com/#json=8DxFWGT66eisYHShw5etd,FTPXATItDi33jkpVtVr54A), + Glazov Sergei - brainstorming, - informtaion research,