Menlo is an AI-powered research intelligence mining platform that automatically collects the latest papers from arXiv, analyzes them with AI, and evaluates their commercial value using the V3 CAV scoring system.
- Auto Mining - Automatically collects latest AI/ML papers from arXiv every 6 hours
- AI Analysis - Intelligent paper analysis using large language models
- V3 CAV Scoring - Four-dimensional scoring system (BIP/TF/CP/RR)
- Bilingual Support - English and Chinese interface
- Responsive Design - Desktop and mobile friendly
| Home | Detail |
|---|---|
 |
 |
- Go 1.21+
- Node.js 18+
- SQLite3
# Clone the repository
git clone https://github.com/your-org/menlo.git
cd menlo
# Configure environment variables
cp .env.example .env
# Edit .env and add your BAILIAN_API_KEY
# Install backend dependencies
go mod download
# Install frontend dependencies
cd frontend && npm install && cd ..
# Build frontend
cd frontend && npm run build && cd ..
# Start the server
go run ./cmd/server/Open your browser and visit http://localhost:8888
| Environment Variable | Description | Default |
|---|---|---|
BAILIAN_API_KEY |
Bailian API key (required) | - |
BAILIAN_MODEL |
Model to use | qwen3.5-plus-2026-02-15 |
SERVER_PORT |
Server port | 8888 |
SCHEDULER_ENABLED |
Enable scheduled mining | true |
MINING_INTERVAL |
Mining interval (cron) | 0 0 */6 * * * |
DB_PATH |
Database path | data/menlo.db |
Menlo uses a four-dimensional scoring system to evaluate paper value:
| Dimension | Weight | Description |
|---|---|---|
| BIP (Business Impact) | 37% | Business impact and application value |
| TF (Technical Feasibility) | 26% | Technical feasibility and implementation difficulty |
| CP (Competitive Position) | 21% | Competitive advantage and market positioning |
| RR (Research Reliability) | 16% | Research reliability and academic quality |
CAV Total Score = BIP(37%) + TF(26%) + CP(21%) + RR(16%)
Core Philosophy: GitHub is a bonus factor (≤5%), not a requirement. Commercial value prioritizes business impact and technical feasibility over code availability.
| Endpoint | Method | Description |
|---|---|---|
/api/intelligence |
GET | Get intelligence list |
/api/intelligence/:id |
GET | Get single intelligence details |
/api/intelligence |
POST | Create intelligence |
/api/intelligence/sync |
POST | Sync external data |
/api/score/v3 |
POST | Calculate V3 CAV score |
/health |
GET | Health check |
# Get intelligence list
curl http://localhost:8888/api/intelligence?page=1&limit=10
# Calculate score
curl -X POST http://localhost:8888/api/score/v3 \
-H "Content-Type: application/json" \
-d '{"title": "...", "abstract": "..."}'- Go + Echo Framework
- SQLite + sqlc
- Cron Scheduling
- React 18
- TypeScript
- Tailwind CSS
- Vite
- Alibaba Cloud Bailian (Qwen)
menlo/
├── cmd/ # CLI applications
│ ├── server/ # API server
│ └── migrate/ # Database migrations
├── internal/ # Internal code
│ ├── domain/ # Domain logic
│ ├── httpapi/ # HTTP API
│ ├── repository/ # Data access layer
│ └── infrastructure/ # External integrations
├── frontend/ # React frontend
├── docs/ # Documentation
├── migrations/ # SQL migrations
└── deployments/ # Deployment configs
# Start backend in development mode
go run ./cmd/server/
# Start frontend in development mode
cd frontend && npm run dev
# Run tests
go test ./...
# Test coverage
go test -cover ./...# Set environment variable
export BAILIAN_API_KEY=your_key
docker-compose -f deployments/docker-compose.yml up -d# Build
go build -o server ./cmd/server/
cd frontend && npm run build
# Start
./server- Bailian AI - LLM qualitative evaluation
- PapersWithCode - SOTA status, benchmarks, datasets
- Semantic Scholar - Citations, FWCI
- GitHub API - Repository metadata
Issues and Pull Requests are welcome!
- arXiv - Open access paper repository
- Alibaba Cloud Bailian - AI model service