Schema

Server Configuration

Describes the environment variables required to run the server.

Name	Required	Description	Default
`DEBUG`	No	Enable debug logging	false
`NODE_ENV`	No	Affects logging behavior (development, test, production)
`OPENAI_ORG_ID`	No	OpenAI organization ID
`OPENAI_API_KEY`	No	OpenAI API key for AI-enhanced tools
`OPENAI_BASE_URL`	No	OpenAI-compatible API endpoint	https://api.openai.com/v1
`OPENAI_PROVIDER`	No	Provider: openai, qwen, azure, anthropic, together	openai
`AMBIANCE_API_KEY`	No	Ambiance cloud API key for cloud features
`AMBIANCE_API_URL`	No	Ambiance cloud API URL	https://api.ambiance.dev
`WORKSPACE_FOLDER`	No	Project workspace path
`AMBIANCE_BASE_DIR`	No	Override working directory
`OPENAI_BASE_MODEL`	No	Primary model for analysis tasks	gpt-5
`OPENAI_MINI_MODEL`	No	Faster model for hints/summaries	gpt-5-mini
`LOCAL_STORAGE_PATH`	No	Custom local storage path	~/.ambiance/embeddings
`EMBEDDING_BATCH_SIZE`	No	Number of texts per embedding batch	32
`USE_LOCAL_EMBEDDINGS`	No	Enable local embedding storage	false
`AMBIANCE_DEVICE_TOKEN`	No	Device identification token
`LOCAL_EMBEDDING_MODEL`	No	Local embedding model when using local embeddings	all-MiniLM-L6-v2
`USING_LOCAL_SERVER_URL`	No	Use local Ambiance server instead of cloud
`EMBEDDING_PARALLEL_MODE`	No	Enable parallel embedding generation	false
`OPENAI_EMBEDDINGS_MODEL`	No	Model for generating embeddings	text-embedding-3-large
`EMBEDDING_MAX_CONCURRENCY`	No	Max concurrent API calls for parallel mode	10
`EMBEDDING_RATE_LIMIT_RETRIES`	No	Max retries for rate limit errors	5
`EMBEDDING_RATE_LIMIT_BASE_DELAY`	No	Base delay for rate limit retries (ms)	1000

Interactive templates invoked by user choice

Name	Description
No prompts

Contextual data attached and managed by the client

Name	Description
No resources

Functions exposed to the LLM to take actions

Name	Description
No tools