CodeFlowIQ is a local-first repository intelligence platform that helps developers understand unfamiliar codebases faster. It indexes a local Git repository or plain folder and turns source code into a readable product experience: runtime stories, repository exploration, API discovery, backend-to-SQL tracing, Azure dependency visibility, and evidence-backed execution understanding.
Built as a serious engineering product rather than a demo dashboard, CodeFlowIQ focuses on a real developer pain point: how do you understand a large enterprise repository without spending days hopping across files, breakpoints, API handlers, DI registrations, repositories, stored procedures, and cloud integrations?
It is especially relevant for:
- Recruiters evaluating product-minded engineering work.
- Interviewers looking for architecture thinking, full-stack execution, and developer-tooling depth.
- Engineering leaders who care about onboarding, maintainability, and local-first analysis.
- Developers trying to answer, "Where does this request actually go?"
CodeFlowIQ is more than code search. It is a developer experience layer on top of static analysis.
- It combines product design, frontend UX, backend architecture, analyzers, indexing, and query orchestration in one cohesive tool.
- It is built around real evidence, not vague summaries.
- It treats readability and trust as product features.
- It supports enterprise-style stacks such as C#, ASP.NET Core, SQL/T-SQL, JavaScript/TypeScript, Angular, React, and Azure signals.
- It is designed to stay local-first, which matters for private repositories and regulated environments.
CodeFlowIQ, code intelligence platform, repository explorer, runtime map, C# code analysis, ASP.NET Core API analysis, SQL dependency analysis, Azure dependency mapping, developer onboarding tool, local code search, static analysis platform, source evidence explorer, execution trace visualizer, enterprise codebase understanding, backend trace debugger
CodeFlowIQ helps a developer move from "I just cloned this repo" to "I understand the important flows, APIs, data touchpoints, and cloud dependencies" through a guided, evidence-backed UI.
Instead of throwing raw search results at the user, it answers high-value engineering questions:
- What is this repository built with?
- Where should I start reading?
- Which APIs are exposed?
- Which backend service actually handles this route?
- Which SQL tables or stored procedures are touched?
- Which Azure services are involved?
- Which parts of the flow are exact, inferred, partial, duplicated, or unresolved?
For recruiters and interviewers, CodeFlowIQ demonstrates:
- Full-stack product engineering across
React,TypeScript,ASP.NET Core,C#, indexing, and query design. - Strong product thinking: complex analysis translated into readable workflows.
- System design maturity: analyzers, background indexing, storage, query handlers, and feature-specific UI modules.
- UX discipline: curated views, drill-down paths, evidence visibility, dark/light theme support, and large-data readability concerns.
- Enterprise empathy: local-first execution, source-backed trust, repository-scale navigation, and onboarding use cases.
- Indexes local Git repositories and plain local folders.
- Detects files, languages, symbols, API routes, SQL objects, Azure service usage, and source relationships.
- Builds readable Runtime Map stories from likely entry points to downstream evidence.
- Provides a full Repository Explorer for browsing files, APIs, backend/data evidence, and cloud signals.
- Traces C# backend execution through controllers, DI handoffs, concrete implementations, base methods, repositories, SQL boundaries, and unresolved calls.
- Connects frontend/API/backend/database/cloud evidence where source relationships support it.
- Keeps analysis local by default.
Large repositories are hard to understand because code search only answers narrow questions. Developers usually need the bigger execution picture:
- Which controller handles this API?
- Which manager or service is resolved by DI?
- Which repository method runs next?
- Which SQL table or stored procedure is touched?
- Which Azure service is used?
- Which evidence is exact, inferred, duplicated, partial, or unresolved?
CodeFlowIQ turns indexed repository evidence into product-shaped workflows for onboarding, debugging, impact analysis, and architecture discovery.
- Runtime Map instead of raw file lists: show the most useful starting stories first.
- Repository Explorer as a central evidence workspace: move from preview to exact drill-down.
- C# backend trace: follow controller to DI to manager to repository to SQL/cloud boundaries.
- Evidence-aware UX: show exact, inferred, partial, duplicate, and unresolved analysis honestly.
- Local-first architecture: no cloud dependency required for the core experience.
- C#
- ASP.NET / ASP.NET Core
- Azure Functions
- SQL / T-SQL
- JavaScript / TypeScript
- React
- Angular
- Azure service usage signals
- Start here: beginner-friendly repository tour, technologies, important folders, data touchpoints, APIs, and Azure dependencies.
- Runtime stories: curated runtime map with recommended stories, all start points, all flows, and source evidence.
- Browse evidence: Repository Explorer for full indexed evidence, details, related context, grouping, and drill-down.
- Repo snapshot: summary of repository structure, technologies, and detected capabilities.
- End-to-end flows: traces feature/API paths across UI, API, backend, database, and cloud evidence.
- C# backend trace: debugger-style backend trace from route/controller into DI, managers, repositories, SQL, and cloud boundaries.
- Backend & data: backend relationships, SQL touchpoints, stored procedures, reads, and writes.
- API endpoints: discovered backend routes and route handlers.
- Cloud services: detected Azure/cloud service references.
- Files indexed: repository file browsing and file-level evidence.
- Settings: API URL, theme, and trace behavior.
CodeFlowIQ/
src/
CodeFlowIQ.Core/ Shared contracts, query DTOs, interfaces
CodeFlowIQ.Data/ SQLite persistence and EF migrations
CodeFlowIQ.Analyzers/ C#, SQL/T-SQL, JS/TS, Angular analyzers
CodeFlowIQ.Indexing/ Indexing services, query handlers, trace resolvers
CodeFlowIQ.Git/ Git workspace detection/support
CodeFlowIQ.Api/ Local ASP.NET Core API
CodeFlowIQ.Cli/ CLI surface
CodeFlowIQ.UI/ React/Vite frontend
tests/
CodeFlowIQ.Tests/ Integration, indexing, and query behavior tests
docs/
assets/screenshots/ README and product screenshots
Install these before running CodeFlowIQ on a new machine:
- .NET SDK compatible with the solution target framework.
- Node.js and npm for the React UI.
- Git is recommended, but target workspaces can also be plain folders.
Check local versions:
dotnet --info
node --version
npm --versionClone the repository:
git clone <your-codeflowiq-repo-url>
cd CodeFlowIQRestore and build the .NET solution:
dotnet restore
dotnet buildInstall UI dependencies:
cd src\CodeFlowIQ.UI
npm.cmd ci
cd ..\..Use npm.cmd on Windows PowerShell. Calling npm directly may resolve to npm.ps1 and hit execution-policy restrictions.
Run the API in one terminal:
cd CodeFlowIQ
$env:CODEFLOWIQ_API_URLS='http://127.0.0.1:5188'
dotnet run --project src\CodeFlowIQ.Api\CodeFlowIQ.Api.csprojRun the UI in a second terminal:
cd CodeFlowIQ\src\CodeFlowIQ.UI
npm.cmd run devOpen the UI:
http://127.0.0.1:5173/
The UI defaults to:
http://127.0.0.1:5188
You can also set the UI API URL before starting Vite:
$env:VITE_CODEFLOWIQ_API_BASE_URL='http://127.0.0.1:5188'
npm.cmd run devInvoke-WebRequest -Uri 'http://127.0.0.1:5188/health' -UseBasicParsingExpected response:
{
"status": "healthy",
"name": "CodeFlowIQ.Api"
}- Start the API and UI.
- Open
http://127.0.0.1:5173/. - Enter a target workspace path, for example:
C:\Users\you\source\SomeEnterpriseRepo
- Click
Index repofor a new workspace. - Watch background indexing progress.
- Click
Load insightsafter indexing completes. - Use Start Here, Runtime Map, Repository Explorer, End-to-End Flows, C# Backend Trace, API Endpoints, Backend & Data, Cloud Services, and Files Indexed to inspect the codebase.
After analyzer changes, refresh or re-index the target workspace so new relationship signals are captured.
Base URL:
http://127.0.0.1:5188
Endpoints:
GET /health
POST /api/workspace/init
POST /api/workspace/sync
GET /api/workspace/status
GET /api/workspace/indexing-status
GET /api/summary
GET /api/overview
GET /api/runtime-flows
GET /api/explorer/*
GET /api/files
GET /api/symbols
GET /api/relationships
GET /api/apis
GET /api/azure
GET /api/flows
GET /api/chains
GET /api/backend
Example Runtime Map query:
$path = [System.Uri]::EscapeDataString('C:\Users\you\source\SomeEnterpriseRepo')
Invoke-RestMethod -Uri "http://127.0.0.1:5188/api/runtime-flows?path=$path&take=10"The API writes local runtime metadata here:
Windows:
%LOCALAPPDATA%\CodeFlowIQ\runtime\api.json
macOS:
~/Library/Application Support/CodeFlowIQ/runtime/api.json
Linux:
~/.local/share/CodeFlowIQ/runtime/api.json
For development, prefer a stable local API port:
$env:CODEFLOWIQ_API_URLS='http://127.0.0.1:5188'Run backend tests:
cd CodeFlowIQ
dotnet testBuild the UI:
cd CodeFlowIQ\src\CodeFlowIQ.UI
npm.cmd run buildIf tests fail because CodeFlowIQ.Api is locking DLLs, stop the running API process and rerun tests.
Find the process:
Get-Process | Where-Object { $_.ProcessName -eq 'CodeFlowIQ.Api' } | Select-Object Id,ProcessName,PathStop it:
Stop-Process -Id <process-id>Check API health:
Invoke-WebRequest -Uri 'http://127.0.0.1:5188/health' -UseBasicParsingIf the API is not running, start it:
$env:CODEFLOWIQ_API_URLS='http://127.0.0.1:5188'
dotnet run --project src\CodeFlowIQ.Api\CodeFlowIQ.Api.csprojIf using a non-default API port, update the UI API field or set:
$env:VITE_CODEFLOWIQ_API_BASE_URL='http://127.0.0.1:<port>'
npm.cmd run devUse npm.cmd:
npm.cmd ci
npm.cmd run dev
npm.cmd run build- Refresh or re-index after analyzer changes.
- Confirm the target workspace path is correct.
- Check Runtime Map quality and coverage.
Detected onlymeans CodeFlowIQ found a start point but did not connect enough downstream evidence.Inferredmeans the connection is useful but not as strong as exact source-backed evidence.- Open the related row in Repository Explorer to inspect the source evidence.
- Keep optional capabilities modular and turn-off-able.
- Prefer deterministic local analysis before LLM or embedding features.
- Do not require Git for core code understanding.
- Keep indexing incremental and mindful of lower-memory machines.
- Keep Runtime Map curated, searchable, and honest about coverage.
- Make Repository Explorer the central evidence workspace.
- Treat C#, SQL, frontend, and cloud traces as stack-specific features before stitching them together.
Near-term priorities:
- Improve C# backend trace resolver quality for DI, keyed services, factories, base classes, repositories, SQL, and cloud handoffs.
- Finish Repository Explorer as the central source evidence workspace.
- Add stronger grouping, filters, pagination, and source-location jump behavior.
- Expand stack-specific tracing for SQL and frontend after the C# trace experience is dependable.
- Add optional local embeddings, documentation indexing, Git history, and LLM-assisted explanation as modular layers.
CodeFlowIQ is released under the MIT License.




