╦═╗┌─┐┌─┐┌─┐╔╗ ┬─┐┬┌┬┐┌─┐┌─┐ ╠╦╝├┤ ├─┘│ │╠╩╗├┬┘│ │││ ┬├┤ ╩╚═└─┘┴ └─┘╚═╝┴└─┴─┴┘└─┘└─┘Cross-repo intelligence CLI
Research, plan, and ship features that span multiple repositories.
You're building a feature that touches 3 repos — a backend API, a frontend app, and a shared library. You need to:
- Understand how auth works across all of them
- Plan which files to create/modify in each repo
- Catch that the backend returns
snake_casebut the frontend expectscamelCase - Find out the event names don't match between producer and consumer
You'd normally spend hours reading code across repos, mentally mapping dependencies, and discovering conflicts mid-implementation.
RepoBridge does this in seconds.
| Feature | Description |
|---|---|
| Ask | Ask questions about your codebase with full cross-repo context. Concise answers with repo/file citations. |
| Plan | Generate detailed implementation plans with per-file tasks, dependency ordering, and cross-cutting concerns. Auto-exports a structured markdown report. |
| Discuss | Multi-phase cross-repo analysis — each repo is analyzed individually, then cross-referenced for conflicts with severity ratings and resolution suggestions. |
| Index | Scans repos for file trees, API routes, exports, events, auth patterns, and env vars. Feeds this as context to AI. |
# From npm (recommended)
npm install -g repobridge
# Or try without installing
npx repobridge
# Or from source
git clone https://github.com/AlameerAshraf/RepoBridge.git
cd RepoBridge
npm install && npm run build && npm link# Launch interactive mode
repobridgeInside the session:
repobridge > init my-platform
repobridge:my-platform > add
repo 1 > /path/to/backend
repo 2 > /path/to/frontend
choice (1/2) > 2
repobridge:my-platform > ask "How does authentication flow from frontend to backend?"
repobridge:my-platform > plan "Add Stripe billing with usage-based pricing"
repobridge:my-platform > discuss "API contract for billing endpoints"
RepoBridge is model-agnostic. Switch providers with one command:
config --provider anthropic # Claude (default)
config --provider openai --model gpt-4o # OpenAI
config --provider deepseek --model deepseek-chat --base-url https://api.deepseek.com/v1
config --provider gemini --model gemini-2.0-flash
config --provider ollama --model llama3.1 # Local, no API key needed
| Provider | Default Model | Auth |
|---|---|---|
anthropic |
claude-sonnet-4-20250514 |
ANTHROPIC_API_KEY |
openai |
gpt-4o |
OPENAI_API_KEY |
gemini |
gemini-2.0-flash |
GEMINI_API_KEY |
ollama |
llama3.1 |
None (local) |
Any OpenAI-compatible API works via --provider openai --base-url <url> (DeepSeek, Groq, Together, Azure, etc.)
| Command | Description |
|---|---|
init <name> |
Create a new project |
projects |
List all projects |
use <name> |
Switch active project |
leave |
Deactivate current project |
delete [name] |
Delete a project (with confirmation) |
status |
Project dashboard — repos, file/route counts, plans, discussions, AI provider |
| Command | Description |
|---|---|
add [path-or-url] |
Guided flow to add repos (minimum 2 required). Supports local paths and GitHub URLs. Auto-indexes on add. |
index |
Re-index all repos in the active project |
| Command | Description |
|---|---|
ask "<question>" |
Ask anything about your repos. Concise, direct answers with key file citations. Session auto-saved with model info. |
plan "<feature>" |
Generate a cross-repo implementation plan. Per-file tasks with details, dependencies, and concerns. Exports markdown report. |
discuss [feature] |
Three-phase cross-repo analysis: per-repo deep analysis → cross-reference for conflicts → validation pass. Each conflict includes severity reasoning and a resolution suggestion. |
| Command | Description |
|---|---|
sessions |
List saved sessions with date, model, and question |
sessions load <id> |
Replay a past session (supports partial ID match) |
sessions delete <id> |
Delete a saved session |
config |
View/change AI provider and model |
Discuss mode is RepoBridge's core feature. It finds integration conflicts before you write code.
repobridge:my-app > discuss "Add real-time notifications"
How it works:
- Analysis phase — Each repo is analyzed individually by the AI, grounded in actual code. The AI examines API contracts, data models, auth patterns, events, and naming conventions.
- Cross-reference phase — All repo analyses are compared in a single pass to identify concrete conflicts with code evidence from both sides.
- Validation phase — If conflicts are found, a final pass validates them, removes false positives, and catches anything missed.
Each conflict includes:
- What's wrong — detailed description citing specific files/endpoints in both repos
- Severity — high/medium/low with reasoning (based on actual impact, not just type)
- Resolution — specific suggested fix
Conflict types detected:
| Type | Example |
|---|---|
endpoint_mismatch |
Backend serves /api/users, frontend calls /users |
field_naming |
Backend returns created_at, frontend expects createdAt |
auth_contract |
Backend expects Bearer token, frontend sends session cookie |
event_mismatch |
Producer emits user.created, consumer listens for user_created |
response_shape |
Backend returns { data: [...] }, frontend expects { results: [...] } |
schema_mismatch |
Backend has email as required, frontend form treats it as optional |
A full markdown report is saved to ~/.repobridge/projects/<name>/discussions/<id>-discussion.md.
Every plan command generates a structured markdown report at ~/.repobridge/projects/<name>/plans/<id>-plan.md:
# Implementation Plan
> **Add Stripe billing with usage-based pricing**
| | |
|---|---|
| **Project** | my-platform |
| **Repos** | 2 |
| **Tasks** | 14 |
| **Blockers** | 3 (1 high) |
## backend
### 1. + CREATE `src/models/Subscription.ts`
Subscription model with Stripe integration
- Fields: id, userId, stripeCustomerId, plan, status, currentPeriodEnd
- Stripe webhook handler for subscription.updated events
> **Depends on:** `src/models/User.ts`
### 2. ~ MODIFY `src/routes/index.ts`
Register billing routes
- Add /api/billing/* route group
- Apply auth middleware to all billing endpoints
...Run repobridge with no arguments to enter a persistent interactive session:
$ repobridge
╦═╗┌─┐┌─┐┌─┐╔╗ ┬─┐┬┌┬┐┌─┐┌─┐
╠╦╝├┤ ├─┘│ │╠╩╗├┬┘│ │││ ┬├┤
╩╚═└─┘┴ └─┘╚═╝┴└─┴─┴┘└─┘└─┘
Interactive mode. Type 'help' for commands, 'exit' to quit.
repobridge > init my-app
repobridge:my-app > add
repobridge:my-app > ask "How does the API handle auth?"
repobridge:my-app > plan "Add user notifications"
repobridge:my-app > discuss
repobridge:my-app > status
repobridge:my-app > exit
All commands work the same as one-off CLI calls. The prompt shows your active project.
Everything is stored locally. No cloud, no telemetry.
~/.repobridge/
├── config.json ← active project + AI provider config
├── repos/ ← cloned GitHub repos
└── projects/
└── <project-name>/
├── config.json ← repos list + metadata
├── index/
│ └── <repo>.json ← indexed repo context
├── sessions/
│ └── <id>.json ← saved Q&A sessions (with model info)
├── plans/
│ ├── <id>.json ← plan data
│ └── <id>-plan.md ← generated markdown report
└── discussions/
├── <id>.json ← discussion results + conflicts
└── <id>-discussion.md ← generated markdown report
src/
├── index.ts ← CLI entry point (commander + REPL)
├── commands/
│ ├── init.ts ← Project creation
│ ├── projects.ts ← List projects
│ ├── use.ts / leave.ts ← Switch/deactivate project
│ ├── add.ts ← Guided repo adding flow
│ ├── index.ts ← Repo indexing
│ ├── ask.ts ← AI Q&A with streaming
│ ├── plan.ts ← Plan generation + markdown export
│ ├── discuss.ts ← Three-phase cross-repo discussion
│ ├── delete.ts ← Project deletion
│ ├── sessions.ts ← Session management (list, load, delete)
│ ├── status.ts ← Project dashboard
│ ├── config.ts ← AI provider configuration
│ └── repl.ts ← Interactive session loop
├── lib/
│ ├── ai.ts ← AI orchestration (ask, plan, discuss prompts)
│ ├── discuss.ts ← Discussion engine (analyze, cross-reference, validate)
│ ├── indexer.ts ← Repo scanning (files, routes, exports, events)
│ ├── storage.ts ← Data types + file I/O
│ ├── prompt.ts ← Shared readline with EOF handling
│ └── providers/
│ ├── base.ts ← LLMProvider interface
│ ├── index.ts ← Provider factory
│ ├── anthropic.ts ← Claude (via SDK)
│ ├── openai.ts ← OpenAI-compatible (native fetch)
│ ├── gemini.ts ← Google Gemini (native fetch)
│ └── ollama.ts ← Ollama local (native fetch)
└── ui/
├── theme.ts ← Layout system, boxes, tables, icons
└── markdown.ts ← Terminal markdown renderer
| Variable | Required | Description |
|---|---|---|
ANTHROPIC_API_KEY |
If using Anthropic | Claude API key |
OPENAI_API_KEY |
If using OpenAI | OpenAI API key |
GEMINI_API_KEY |
If using Gemini | Google AI API key |
Ollama requires no API key. Keys can also be set via config --api-key.
- Node.js >= 18
- npm
npm testCI runs build + tests on every push and PR to main.
MIT
Stop reading code across 5 repos. Let them talk to each other.