deep-thinker
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
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 | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| thinkB | Add a thought to the cognitive graph using the current strategy. Supports sequential, dialectic, parallel, analogical, and abductive reasoning strategies. Each thought becomes a node in a DAG with confidence scoring, edges, and metadata. |
| evaluateC | Evaluate the thinking process: score confidence, generate critiques, and assess overall graph health. Provides detailed analysis of weak spots and strong reasoning paths. |
| metacogA | Metacognitive operations: view the current thinking state, get strategy suggestions, switch strategies, and receive guidance on improving reasoning. The system automatically detects stuck states and recommends actions. |
| graphC | Query and visualize the thought graph. View the DAG structure, find paths, inspect branches, and get statistics. |
| pruneB | Prune and optimize the thought graph. Remove dead ends, consolidate redundant branches, and optimize reasoning paths. Helps maintain graph efficiency during deep reasoning. |
| resetA | Reset the thought graph and metacognitive state. Start a fresh reasoning session. |
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 6 tools
Each tool has a clearly distinct purpose with no overlap: evaluate analyzes thinking quality, graph visualizes structure, metacog manages reasoning strategy, prune optimizes the graph, reset clears state, and think adds new thoughts. The descriptions reinforce unique roles, making misselection unlikely.
All tool names follow a consistent verb-based pattern (evaluate, graph, metacog, prune, reset, think) with no mixing of conventions. The names are concise, descriptive, and uniformly styled, making them predictable and easy to understand.
With 6 tools, the set is well-scoped for a deep reasoning system, covering core operations like adding thoughts, analyzing, visualizing, optimizing, managing strategy, and resetting. Each tool earns its place without feeling excessive or insufficient for the domain.
The toolset provides complete coverage for deep reasoning workflows: think for creation, evaluate for analysis, graph for inspection, metacog for strategy, prune for optimization, and reset for cleanup. There are no obvious gaps, enabling agents to handle the full lifecycle of cognitive processing.