Server Configuration
Describes the environment variables required to run the server.
Name | Required | Description | Default |
---|---|---|---|
QWEN_API_KEY | No | Your Qwen API key | |
CIPHER_PG_URL | No | PostgreSQL connection URL for chat history | |
GEMINI_API_KEY | No | Your Gemini API key | |
OPENAI_API_KEY | No | Your OpenAI API key | |
REDACT_SECRETS | No | Redact secrets in logs | true |
MCP_SERVER_MODE | No | MCP server mode | aggregator |
CIPHER_LOG_LEVEL | No | Logging level: error, warn, info, debug, silly | info |
VECTOR_STORE_URL | No | Vector store URL (e.g., https://your-cluster.qdrant.io) | |
ANTHROPIC_API_KEY | No | Your Anthropic API key | |
AWS_ACCESS_KEY_ID | No | AWS access key ID for Bedrock | |
VECTOR_STORE_TYPE | No | Vector store type: qdrant, milvus, or in-memory | in-memory |
AWS_DEFAULT_REGION | No | AWS default region | us-east-1 |
SEARCH_MEMORY_TYPE | No | Memory search type: knowledge, reflection, both | knowledge |
USE_WORKSPACE_MEMORY | No | Enable workspace memory | true |
VECTOR_STORE_API_KEY | No | Vector store API key | |
AWS_SECRET_ACCESS_KEY | No | AWS secret access key for Bedrock | |
VECTOR_STORE_DISTANCE | No | Vector store distance metric: Cosine, Euclidean, Dot, Manhattan | Cosine |
VECTOR_STORE_DIMENSION | No | Vector store dimension | 1536 |
VECTOR_STORE_MAX_VECTORS | No | Maximum number of vectors in store | 10000 |
DISABLE_REFLECTION_MEMORY | No | Disable reflection memory | true |
WORKSPACE_VECTOR_STORE_COLLECTION | No | Workspace memory collection name | workspace_memory |
Schema
Prompts
Interactive templates invoked by user choice
Name | Description |
---|---|
No prompts |
Resources
Contextual data attached and managed by the client
Name | Description |
---|---|
No resources |
Tools
Functions exposed to the LLM to take actions
Name | Description |
---|---|
No tools |