Cognitive Exoskeleton MCP Server
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| LLM_API_KEY | No | API key for the LLM provider | EMPTY |
| LLM_API_BASE | No | OpenAI-compatible API base URL | http://127.0.0.1:8000/v1 |
| LLM_MODEL_NAME | No | Model name to use | hy3 |
| COGNITIVE_DB_PATH | No | SQLite database file path | ./cognitive.db |
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 |
|---|---|
| ingest_noteA | Extract knowledge entities and relationships from a note or text, and store them in the personal knowledge graph. Accepts either raw text content or a file path to a Markdown/text file. |
| query_mindA | Answer a question using your personal knowledge graph. Retrieves relevant entities and relationships, then uses the LLM to reason over them. Supports shallow (1-hop) and deep (2-hop) retrieval modes. |
| recall_contextA | While you are writing, surface related notes and ideas from your knowledge graph that you may have forgotten. Helps connect current work with past knowledge. |
| discover_connectionsA | Discover hidden, non-obvious connections between knowledge entities from different domains. If a topic is given, finds connections related to that topic. Without a topic, scans the whole graph for cross-domain bridges. |
| detect_blindspotsA | Analyze a topic's coverage in your knowledge graph and identify blindspots, contradictions, and missing perspectives. Helps you understand what you don't know about a subject. |
| analyze_cognitive_topologyA | Analyze the overall structure of your knowledge graph. Generates a 'cognitive portrait' showing knowledge islands, bridge concepts, dense/sparse regions, and recommendations for improving knowledge connectivity. |
| trace_concept_evolutionA | Trace how your understanding of a concept has evolved over time. Shows a timeline of belief changes, key turning points, and what triggered each shift in understanding. |
| spark_serendipityA | Deliberately spark creative inspiration by colliding concepts from two different knowledge domains. Like a digital serendipity engine — find unexpected connections that neither domain has explored alone. |
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 8 tools
Each tool has a distinct primary purpose, but discover_connections and spark_serendipity both involve cross-domain connections, and detect_blindspots and analyze_cognitive_topology both analyze the graph. Descriptions help differentiate them, but there's slight potential for misselection.
All eight tools follow a consistent verb_noun pattern with snake_case (e.g., ingest_note, trace_concept_evolution). No mixed conventions or vague verbs.
Eight tools is well-scoped for a personal knowledge graph assistant, covering a range of cognitive operations without excess. Each tool serves a clear function.
The set covers ingestion, querying, recall, and analysis well, but lacks explicit edit/delete or raw listing of knowledge entities. This is a minor gap since the focus is on cognitive workflows rather than full CRUD.