Expand description
Configuration management for Terraphim AI.
Provides role-based configuration where each Role describes a user profile with
a set of Haystacks (data sources), a relevance function, and optional LLM settings.
§Loading Priority
- Explicit path via
TERRAPHIM_CONFIGenvironment variable - Saved config retrieved from the persistence layer
- Hard-coded defaults in
terraphim_server/default/
§Key Types
Config– top-level configuration holding all rolesRole– user profile with haystacks, relevance function, and themeHaystack– a data source descriptor (path, service type, extra parameters)ServiceType– enum of supported haystack backends
Modules§
- llm_
router - LLM Router Configuration Types
- project
Structs§
- Config
- The Terraphim config is the main configuration for terraphim
- Config
Builder - Builder, which allows to create a new
Config - Config
State - Config state can be updated using the API or Atomic Server
- Haystack
- A haystack is a collection of documents that can be indexed and searched
- Knowledge
Graph - A knowledge graph is the collection of documents which were indexed using a specific service
- Knowledge
Graph Local - Local knowledge-graph source: an input type paired with a filesystem path.
- Role
- A role is a collection of settings for a specific user
Enums§
- Config
Id - Distinguishes how the configuration is deployed: as a background server, a desktop application, or compiled-in (WASM/library) mode.
- Service
Type - The service used for indexing documents
- Terraphim
Config Error - Errors arising from loading, validating, or persisting Terraphim configuration.
Functions§
- expand_
path - Expand shell-like variables in a path string.
Type Aliases§
- Result
- Convenience alias for
Result<T, TerraphimConfigError>used throughout this crate.