Multi-Agent Sequential Thinking with Ollama — Active validation layer for the MCP sequential-thinking server.
Quick Install
Copy the config for your editor. Some servers may need additional setup — check the README.
Claude Desktop
Claude Code
Cursor
Add to claude_desktop_config.json:
{
"mcpServers": {
"mast-ollama": {
"command": "uvx",
"args": [
"mast-ollama"
]
}
}
}📋 Copy
Run in terminal:
claude mcp add mast-ollama uvx mast-ollama📋 Copy
Add to .cursor/mcp.json:
{
"mcpServers": {
"mast-ollama": {
"command": "uvx",
"args": [
"mast-ollama"
]
}
}
}📋 Copy
Or install with pip: pip install mast-ollama
README Excerpt
**Multi-Agent Sequential Thinking with Ollama** — Active validation layer for the [MCP sequential-thinking](https://github.com/modelcontextprotocol/servers/tree/main/src/sequentialthinking) server. Drop-in Python replacement that challenges each reasoning step with local or cloud Ollama models before returning the result to the calling LLM.
Tools (20)
ACTOR_CRITIC_MAX_ROUNDS BRAINSTORM_MODELS BRAINSTORM_SYNTH_MODEL CRITIC_MODEL DEBONO_BLACK_MODEL DEBONO_BLUE_CLOSE_MODEL DEBONO_BLUE_OPEN_MODEL DEBONO_GREEN_MODEL DEBONO_RED_MODEL DEBONO_SKIP_RED DEBONO_WHITE_MODEL DEBONO_YELLOW_MODEL DISABLE_THOUGHT_LOGGING JUDGE_MODEL KALMAN_ACCEPT_THRESHOLD KALMAN_P_THRESHOLD KALMAN_SCORER_MODELS MAST_CACHE_TTL_S MAST_COLOR_THOUGHTS MAST_FORMAT_MODE
Topics
agents ai llm mcp model-context-protocol multi-agent ollama python sequential-thinking