Codebase intelligence for AI and humans: code health scores, auto-generated docs, git analytics, dead code detection, and architectural decisions via MCP.
Python aicode-complexitycode-healthcode-intelligencecode-qualitydead-codedeveloper-toolsdocumentationgit-analyticsmcpopen-sourcepythonrefactoringstatic-analysistechnical-debt
# repowise Environment Variables
# Copy this file to .env and fill in your values.
# Never commit .env to version control.
# ---------------------------------------------------------------------------
# LLM Provider API Keys (set at least one)
# ---------------------------------------------------------------------------
# Anthropic (recommended for best results)
ANTHROPIC_API_KEY=your_anthropic_api_key_here
# OpenAI or an OpenAI-compatible gateway such as 9router
OPENAI_API_KEY=your_openai_api_key_here
# OPENAI_BASE_URL=http://localhost:20128/v1
# DeepSeek
DEEPSEEK_API_KEY=your_deepseek_api_key_here
# DEEPSEEK_BASE_URL=https://api.deepseek.com
# Kimi
KIMI_API_KEY=your_kimi_api_key_here
# KIMI_BASE_URL=https://api.kimi.com/coding/v1
# Eden AI (many vendors via one OpenAI-compatible key; EU-headquartered)
EDENAI_API_KEY=your_edenai_api_key_here
# Set the EU endpoint for data residency / GDPR-sensitive workloads:
# EDENAI_BASE_URL=https://api.eu.edenai.run/v3
# Ollama — no API key needed, but set base URL if non-default
# OLLAMA_BASE_URL=http://localhost:11434
# LiteLLM proxy (for Together AI, Replicate, Groq, etc.)
# LITELLM_API_KEY=your_litellm_key_here
# LITELLM_BASE_URL=http://localhost:4000
# ---------------------------------------------------------------------------
# Database
# ---------------------------------------------------------------------------
# SQLite (default, no setup required)
DATABASE_URL=sqlite+aiosqlite:///./repowise.db
# PostgreSQL (optional, for production deployments)
# DATABASE_URL=postgresql+asyncpg://user:password@localhost/repowise
# ---------------------------------------------------------------------------
# Storage Paths
# ---------------------------------------------------------------------------
# LanceDB vector store path (used when DATABASE_URL is SQLite)
# When DATABASE_URL is PostgreSQL, pgvector is used instead — no separate path needed
LANCEDB_PATH=./data/lancedb
# Dependency graph storage path
GRAPH_PATH=./data/graphs
# ---------------------------------------------------------------------------
# Server
# ---------------------------------------------------------------------------
# Optional API key to protect repowise API endpoints
# Leave empty for fully open access (suitable for local use)
REPOWISE_API_KEY=
# Server bind address and port
REPOWISE_HOST=0.0.0.0
REPOWISE_PORT=7337
# MCP server port (SSE transport)
REPOWISE_MCP_PORT=7338
# ---------------------------------------------------------------------------
# Generation Defaults
# ---------------------------------------------------------------------------
# Default provider: anthropic | openai | ollama | litellm
REPOWISE_DEFAULT_PROVIDER=anthropic
# Default model (depends on provider)
REPOWISE_DEFAULT_MODEL=claude-sonnet-4-6
# Max concurrent LLM calls during generation
REPOWISE_MAX_CONCURRENT=5
# Reasoning mode for supported docs-generation providers: auto | off | minimal
# auto = provider default; off = disable Qwen3/OpenRouter effort reasoning when supported;
# minimal = request the lowest supported OpenAI/OpenRouter reasoning effort.
# Unsupported explicit modes fail before an API call.
REPOWISE_REASONING=auto
# Max pages to regenerate per maintenance run
REPOWISE_CASCADE_BUDGET=30
# ---------------------------------------------------------------------------
# Development
# ---------------------------------------------------------------------------
# Set to "true" to enable debug logging
REPOWISE_DEBUG=false
# Log level: DEBUG | INFO | WARNING | ERROR
REPOWISE_LOG_LEVEL=INFO