knowledge-to-action-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| OPENAI_API_KEY | No | OpenAI API key required if embeddings or planning are enabled. | |
| PLANNING_MODEL | No | Planning model name. | gpt-4.1-mini |
| EMBEDDING_MODEL | No | Embedding model name. | text-embedding-3-small |
| PLANNING_ENABLED | No | Enable optional planning capabilities. | false |
| EXECUTION_ENABLED | No | Enable workspace inspection tools. | false |
| PLANNING_PROVIDER | No | Planning provider. | openai |
| EMBEDDINGS_ENABLED | No | Enable optional embeddings retrieval. | false |
| EMBEDDING_PROVIDER | No | Embedding provider. | openai |
| OBSIDIAN_VAULT_ROOT | Yes | Path to the Obsidian vault root directory. | |
| EXECUTION_TIMEOUT_MS | No | Timeout for execution tools. | 5000 |
| EMBEDDING_SQLITE_PATH | No | Path to SQLite index file for embeddings. | .knowledge-to-action-mcp/index.sqlite |
| EXECUTION_CAPABILITIES | No | Comma-separated list of execution capabilities. | workspace.search,workspace.inspect,workspace.git_status |
| EXECUTION_MAX_OUTPUT_BYTES | No | Maximum output bytes for execution tools. | 32768 |
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": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| vault.list_notesA | List notes in the vault using deterministic ordering. |
| vault.read_noteB | Read one note from the vault and return stable hash metadata. |
| vault.search_notesB | Search note contents in deterministic order. |
| vault.get_metadataB | Return frontmatter and metadata from a note. |
| graph.buildA | Build the knowledge graph from the vault. Parses all wikilinks, tags, and computes backlinks. Must be called before using other graph/context tools. |
| graph.get_neighborsA | Get neighbor nodes of a note in the knowledge graph (BFS traversal). Includes outLinks, backLinks, and shared-tag neighbors. Graph must be built first. |
| graph.get_backlinksA | Get all notes that link to the specified note. Graph must be built first. |
| context.gatherC | Gather related context for a note using the knowledge graph. Returns scored related notes with snippets, ranked by relationship strength. |
| embeddings.index_vaultC | Chunk vault notes and persist optional embeddings to a local SQLite index. |
| context.retrieveC | Retrieve notes using lexical, graph, and optional embedding-based reranking. |
| context.bundle_for_agentC | Build an agent-ready context packet from a note and related vault context. |
| action.plan_from_noteB | Create a preview-only action plan from a note and nearby context. |
| action.handoff_to_repoC | Connect note context to a related workspace using read-only inspection commands. |
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 13 tools
Each tool has a clearly distinct purpose within its category (actions, context, embeddings, graph, vault). No overlapping functionality detected; even tools within the same category (e.g., context.gather and context.retrieve) have different retrieval methods.
All tool names follow a consistent pattern: category.snake_case_action (e.g., action.handoff_to_repo, vault.list_notes). The naming is descriptive and predictable, with no mixed conventions.
13 tools is well within the optimal range for a server of this scope. The count covers all necessary operations for knowledge management and action planning without unnecessary bloat.
The tool surface covers core workflows: graph building, context retrieval, note reading/searching, embedding indexing, and action planning. Missing tools for note creation or modification are acceptable given the read-analysis-plan focus, but a minor gap exists for editing vault content.