Skip to main content
Glama

Hive Mind 🐝

Swarm intelligence for AI agents β€” collective decision-making through weighted consensus voting.

Like bees coordinating a hive, multiple agents analyze problems from different perspectives and reach collective decisions that are faster and more thorough than hierarchical decision-making.

How It Works

  1. Create a Decision β€” Post a question for the swarm

  2. Agents Vote β€” Each agent analyzes from their perspective (financial, legal, technical, market, risk, strategic)

  3. Consensus Emerges β€” Votes are weighted by confidence and expertise

  4. Decision Made β€” The collective recommendation with confidence score

Related MCP server: Claude-Flow

Installation

pip install hive-mind-mcp-server
{"mcpServers": {"hive": {"command": "uvx", "args": ["hive-mind-mcp-server"]}}}

Tools

Tool

Description

create_decision

Post a question for the swarm

cast_vote

Vote with recommendation, reasoning and confidence

get_consensus

See the collective decision with breakdown

close_decision

Lock and finalize a decision

list_decisions

Browse all open/closed decisions

get_agent_expertise

View an agent's expertise profile

Enterprise Use Case

Instead of slow management hierarchies:

  • 5 specialist agents analyze a business decision simultaneously

  • Financial agent checks ROI, legal agent checks compliance, technical agent checks feasibility

  • Consensus in seconds instead of weeks of meetings

  • Full audit trail of reasoning and confidence levels

Network Effect

More agents voting β†’ better expertise data β†’ more accurate weighting β†’ better decisions β†’ more agents using the system.


More MCP Servers by AiAgentKarl

Category

Servers

πŸ”— Blockchain

Solana

🌍 Data

Weather Β· Germany Β· Agriculture Β· Space Β· Aviation Β· EU Companies

πŸ”’ Security

Cybersecurity Β· Policy Gateway Β· Audit Trail

πŸ€– Agent Infra

Memory Β· Directory Β· Hub Β· Reputation

πŸ”¬ Research

Academic Β· LLM Benchmark Β· Legal

β†’ Full catalog (40+ servers)

License

MIT

Available Tools

6 tools
cast_voteA

Cast a vote on an open decision.

Each agent votes from their perspective with a recommendation, reasoning and confidence level. Votes are weighted by expertise.

Args: decision_id: The decision to vote on agent_id: Your agent identifier recommendation: Your recommendation (e.g. "approve", "reject", "modify", or free text) reasoning: Why you recommend this (1-3 sentences) confidence: How confident you are (0.0 to 1.0) perspective: Your analysis angle (financial, technical, legal, market, risk, strategic) expertise_area: Your area of expertise for weighting

Inputs: decision_id (required), agent_id (required), recommendation (required), reasoning, +3 more.

close_decisionA

Close a decision and record the final outcome.

Locks the decision from further votes and records what was decided.

Args: decision_id: The decision to close final_decision: The final decision taken (optional, uses consensus if empty)

Inputs: decision_id (required), final_decision.

create_decisionA

Create a new decision for the swarm to evaluate.

Multiple agents can then vote on this decision from different perspectives (financial, legal, technical, market, risk). The hive mind aggregates votes into a collective decision.

Args: question: The decision question (e.g. "Should we launch product X?") context: Background information for the decision decision_id: Custom ID (auto-generated if empty) created_by: Who initiated the decision

Inputs: question (required), context, decision_id, created_by.

get_agent_expertiseC

Get an agent's expertise profile across all areas.

Shows which areas an agent has voted on and their expertise scores.

Args: agent_id: The agent to look up

Inputs: agent_id (required).

get_consensusA

Get the current consensus status for a decision.

Aggregates all votes, weights by expertise and confidence, and produces a collective decision with confidence score.

Args: decision_id: The decision to analyze

Inputs: decision_id (required).

list_decisionsA

List all decisions in the hive mind.

Args: status: Filter by status ("open", "closed", "all") limit: Max results (default: 20)

Inputs: status, limit.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A3.9/5.0

    Scored across 6 tools

    Disambiguation5/5

    Each tool has a distinct and clear purpose: creating, voting on, closing decisions, getting consensus, listing decisions, and querying agent expertise. No overlap.

    Naming Consistency5/5

    All tool names follow a consistent verb_noun pattern in snake_case (e.g., cast_vote, create_decision, list_decisions).

    Tool Count5/5

    With 6 tools, the server is well-scoped for a decision-making system, covering the core lifecycle without excess or deficiency.

    Completeness5/5

    The tool surface covers the full decision lifecycle: creation, voting, consensus, closing, listing, and agent expertise. No obvious gaps.

    Maintenance

    ActivityInactive
    ResponsivenessNo recent community issues

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      A
      quality
      D
      maintenance
      Enables agentic coordination by connecting humans and AI agents through group messaging, project tracking, and milestone management. It provides tools for consensus voting, progress checkpoints, and multi-session collaboration across various agentic platforms.
      32
      1
      MIT
    • A
      license
      Not graded
      quality
      D
      maintenance
      Multi-agent AI orchestration platform for Claude Code that enables deploying and coordinating specialized agents in swarms with self-learning, fault-tolerant consensus, and enterprise security.
      2
      MIT
    • F
      license
      Not graded
      quality
      Not graded
      maintenance
      Aggregates multiple divergent LLM reasoning trajectories through majority voting and consensus scoring to produce high-confidence deterministic outputs without additional model calls. Plug into any MCP-compliant client to resolve agent tasks with sub-millisecond, zero-dependency execution.
      -