LINZA-MCP
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| LINZA_VAULT | Yes | Path to the Markdown folder (vault). | |
| LINZA_LANGUAGE | No | Language for guide_next_steps: auto, ru, en. | |
| LINZA_EMBED_KEY | No | Optional API key for OpenAI-compatible embeddings. | |
| LINZA_EMBED_URL | Yes | URL for the embeddings API. | |
| LINZA_EMBED_MODEL | Yes | Model name for embeddings. | |
| LINZA_TOOL_SURFACE | No | Tool surface: default or advanced. | |
| LINZA_EMBED_PROVIDER | Yes | Embedding provider: lmstudio, openai, ollama. | |
| LINZA_DEFAULT_PROFILE | No | Default search profile (default: general). | |
| LINZA_BRIDGE_THRESHOLD | No | Semantic bridge threshold (default: 0.55). | |
| LINZA_MAX_BRIDGE_PAIRS | No | Max pairs for bridge computation (default: 1000000, 0 disables). |
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
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| index_allA | Rebuild the LINZA sidecar index for the configured Markdown vault. Use after first setup or after notes change; writes only .linza/linza.db and does not modify note bodies. |
| searchA | Find relevant indexed Markdown notes by semantic/lexical query. Use before read_file when the path is unknown; may record search history in the LINZA sidecar. |
| read_fileA | Read one vault-relative Markdown file exactly as stored. Use after search or when the path is known; does not write files or sidecar state. |
| get_statsA | Return quick LINZA sidecar status: indexed file count, profiles, semantic bridges, and active profile. Use for health checks. |
| scan_vaultA | Run a read-only vault diagnostic over files, links, metadata, and LINZA setup. Use at first contact or when deciding what to fix next. |
| guide_next_stepsA | Read the current LINZA state and recommend the next safe onboarding/review step in plain language. This is the navigator for agents and users; it does not write files. |
| agent_workspaceA | Main LINZA workflow facade. Choose one action to inspect the workspace, ingest artifacts, review/apply supervised items, explain connections, search memory, export context, or run diagnostics. Safe actions are read-only; apply actions preview by default with dry_run=true. |
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
The tools are mostly distinct, but agent_workspace lists actions that overlap with search, read_file, and scan_vault, causing ambiguity about which tool to use for those tasks.
Most tools follow a verb_noun pattern (get_stats, read_file, scan_vault, guide_next_steps), but index_all, search, and agent_workspace deviate, mixing verb-only and noun_noun styles.
7 tools cover the core workflows of indexing, searching, reading, diagnostics, status, and guidance without being too few or too many.
The set lacks write operations (create, update, delete notes), and no tool for managing profiles or semantic bridges directly, leaving significant gaps in CRUD coverage.