dnomia-knowledge
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| DNOMIA_KNOWLEDGE_DB | No | Database path | ~/.local/share/dnomia-knowledge/knowledge.db |
| DNOMIA_KNOWLEDGE_PROJECT | Yes | Default project for MCP search |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| searchA | Hybrid semantic + keyword search across project knowledge. Args: query: Search query text domain: Filter by "all", "code", or "content" project: Project ID (default: DNOMIA_KNOWLEDGE_PROJECT env var) cross: If True, also search linked projects limit: Maximum results to return language: Filter by language (e.g. "python", "typescript") file_pattern: Filter by file path pattern (e.g. "auth", "models.py") show_content: If True, show full chunk content instead of truncated snippet |
| index_projectB | Index or reindex a project's content and code. Args: path: Absolute path to project root (must contain .md/.mdx files) incremental: If True, only reindex changed files (default) |
| project_infoA | List registered projects with stats. Args: project: Specific project ID, or None for all projects |
| graph_queryA | Query the knowledge graph. Args: chunk_id: Chunk ID to start traversal from (required for neighbors mode) project: Project ID (required for communities mode) mode: "neighbors" for BFS traversal, "communities" for Louvain groups depth: BFS depth for neighbors mode (1-3) |
| read_fileA | Smart file reading with index awareness. If the file is indexed and large, returns relevant chunks instead of full content. Falls back to raw file reading for non-indexed files. Args: file_path: Absolute path to the file to read query: Optional search query to find relevant sections in large files project: Project ID (default: auto-detect from file path) |
| fetch_and_indexA | Fetch URL content, convert to text, and index for searching. After indexing, the content is searchable via the search tool. Args: url: URL to fetch and index project: Project ID to store under (default: derived from URL domain) |
| lifecycleA | Inspect or mutate schema v4 lifecycle state for a chunk. Args: chunk_id: Target chunk id action: info | reinforce | supersede | restore new_id: Chunk replacing the target (required when action=supersede) delta: Confidence delta for manual reinforcement (default 0.1) reason: Optional free-text reason stored on the event |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 7 tools
Each tool targets a clearly distinct function: project listing, search, local indexing, graph traversal, file reading, URL ingestion, and lifecycle management. There is no meaningful overlap in purpose even though some tools touch related data.
Tool names mix conventions: index_project and read_file are verb_noun, project_info and graph_query are noun_verb, search is a bare verb, and lifecycle is a noun only. While all names are readable and snake_case, there is no predictable pattern across the set.
Seven tools is a well-scoped size for a knowledge management server, covering indexing, retrieval, graph exploration, and state management without unnecessary redundancy. Each tool earns its place.
The core knowledge lifecycle is covered: ingestion via index_project/fetch_and_index, retrieval via search/read_file, graph access, and state changes via lifecycle. A notable gap is the absence of any deletion/removal tool for projects, chunks, or indexed content.