# Thoughtbox (verified)

Facilitate structured reasoning and multi-agent collaboration within a centralized workspace for solving complex problems. Design, validate, and orchestrate implementation specifications using integr…

## Quick Start

```bash
# Connect this server (installs CLI if needed)
npx -y @smithery/cli@latest mcp add Kastalien-Research/thoughtbox

# Browse available tools
npx -y @smithery/cli@latest tool list Kastalien-Research/thoughtbox

# Get full schema for a tool
npx -y @smithery/cli@latest tool get Kastalien-Research/thoughtbox thoughtbox_gateway

# Call a tool
npx -y @smithery/cli@latest tool call Kastalien-Research/thoughtbox thoughtbox_gateway '{}'
```

## Direct MCP Connection

Endpoint: `https://thoughtbox--kastalien-research.run.tools`

**Required config:**
- `x-api-key` (header) — Thoughtbox is currently free to use with registration. sign up at https://www.the-thoughtbox.dev/

## Tools (3)

- `thoughtbox_gateway` — Single entry point for all Thoughtbox operations.
- `observability_gateway` — Query system observability data including metrics, health status, active sessions, and alerts. No session initializatio…
- `thoughtbox_hub` — Multi-agent collaboration hub for coordinated reasoning.

```bash
# Get full input/output schema for a tool
npx -y @smithery/cli@latest tool get Kastalien-Research/thoughtbox <tool-name>
```

## Resources

- `thoughtbox://init` — START HERE FIRST: Read this resource before using any Thoughtbox tools. Initializes session context and loads previous …
- `system://status` — Health snapshot of the notebook server
- `thoughtbox://notebook/operations` — Complete catalog of notebook operations with schemas and examples
- `thoughtbox://session/operations` — Complete catalog of session operations with schemas and examples
- `thoughtbox://gateway/operations` — Complete catalog of gateway operations (thought, read_thoughts, get_structure, cipher, deep_analysis)
- `thoughtbox://init/operations` — Complete catalog of init operations (get_state, list_sessions, navigate, load_context, start_new, list_roots, bind_root)
- `thoughtbox://knowledge/operations` — Complete catalog of knowledge graph operations with schemas and examples
- `thoughtbox://hub/operations` — Complete catalog of all 27 hub operations with stage metadata and vocabulary
- `thoughtbox://patterns-cookbook` — Guide to core reasoning patterns for thoughtbox tool
- `thoughtbox://architecture` — Interactive notebook explaining Thoughtbox MCP server architecture and implementation patterns
- `thoughtbox://cipher` — Token-efficient notation system for long reasoning chains
- `thoughtbox://session-analysis-guide` — Process guide for qualitative analysis of reasoning sessions (key moments → extract learnings)
- `thoughtbox://guidance/parallel-verification` — Workflow for parallel hypothesis exploration using Thoughtbox branching
- `thoughtbox://prompts/evolution-check` — Check which prior thoughts should be updated when a new insight is added. Same content as evolution-check prompt.
- `thoughtbox://prompts/subagent-summarize` — Context isolation pattern for retrieving sessions. Same content as subagent-summarize prompt.
- `thoughtbox://tests/thoughtbox` — Behavioral tests for the thoughtbox thinking tool (15 tests covering forward/backward thinking, branching, revisions, l…
- `thoughtbox://tests/notebook` — Behavioral tests for the notebook literate programming tool (8 tests covering creation, cells, execution, export)
- `thoughtbox://knowledge/stats` — Entity and relation counts for the knowledge graph
- `thoughtbox://tests/mental-models` — Behavioral tests for the mental_models structured reasoning tool (6 tests covering discovery, retrieval, capability gra…
- `thoughtbox://tests/memory` — Behavioral tests for the knowledge/memory toolhost (12 tests covering patterns, scratchpad, persistence)
- `thoughtbox://mental-models/operations` — Complete catalog of mental models, tags, and operations
- `thoughtbox://loops/catalog` — Complete catalog of OODA loop building blocks with metadata, classification, and composition rules
- `thoughtbox://loops/analytics/refresh` — Trigger immediate aggregation of loop usage metrics and return updated statistics
- `thoughtbox://mental-models` — Root directory: 15 models across 9 tags
- `thoughtbox://mental-models/debugging` — Finding and fixing issues in code, logic, or systems (2 models)
- `thoughtbox://mental-models/planning` — Breaking down work, sequencing tasks, project organization (6 models)
- `thoughtbox://mental-models/decision-making` — Choosing between options under uncertainty (4 models)
- `thoughtbox://mental-models/risk-analysis` — Identifying what could go wrong and how to prevent it (3 models)
- `thoughtbox://mental-models/estimation` — Making reasonable guesses with limited information (1 models)
- `thoughtbox://mental-models/prioritization` — Deciding what to do first, resource allocation (3 models)
- `thoughtbox://mental-models/communication` — Explaining clearly to humans, documentation (2 models)
- `thoughtbox://mental-models/architecture` — System design, component relationships, structure (3 models)
- `thoughtbox://mental-models/validation` — Checking assumptions, testing hypotheses, verification (4 models)

## Prompts (12)

- `list_mcp_assets` — Overview of all MCP capabilities, tools, resources, and quickstart guide
- `interleaved-thinking` (task, thoughts_limit, clear_folder) — Use this Thoughtbox server as a reasoning workspace to alternate between internal reasoning steps and external tool/act…
- `subagent-summarize` (request) — Get instructions for using Claude Code's Task tool to retrieve and summarize Thoughtbox sessions with context isolation…
- `evolution-check` (newThought, sessionId, priorThoughts) — Get instructions for checking which prior thoughts should be updated when a new insight is added. Uses sub-agent patter…
- `test-thoughtbox` — Behavioral tests for the thoughtbox thinking tool (15 tests covering forward/backward thinking, branching, revisions, l…
- `test-notebook` — Behavioral tests for the notebook literate programming tool (8 tests covering creation, cells, execution, export)
- `test-mental-models` — Behavioral tests for the mental_models structured reasoning tool (6 tests covering discovery, retrieval, capability gra…
- `test-memory` — Behavioral tests for the knowledge/memory toolhost (12 tests covering patterns, scratchpad, persistence)
- `spec-designer` (prompt, output_folder, depth, max_specs, plan_only) — Design and produce implementation specifications through structured cognitive loops. Creates specs from prompts using O…
- `spec-validator` (spec_path, strict, deep, report_only) — Systematically validate specification documents against current codebase and project architecture. Identifies gaps, con…
- `spec-orchestrator` (spec_folder, budget, max_iterations, plan_only) — Coordinate implementation of multiple specification documents from a folder. Manages dependencies, tracks progress, and…
- `specification-suite` (prompt_or_spec_path, output_folder, depth, budget, plan_only, skip_design, skip_validation) — Chain the design → validate → orchestrate lifecycle into one command. Moves from blank prompt to implemented, validated…