hive-mind-advanced
Advanced Hive Mind collective intelligence system for queen-led multi-agent coordination with consensus mechanisms and persistent memory
Install / Use
npx skills add foryourhealth111-pixel/Vibe-Skills --skill hive-mind-advancedInstalls into whichever agent you are using.
SKILL.md
Installable skill definition
Quality Score
Category
Development & EngineeringSupported Platforms
Tags
Our assessment of hive-mind-advanced
hive-mind-advanced scores 95/100 on our quality scale, 305th of 4,700 Development & Engineering skills we index (top 7%).
Its SKILL.md is 16 KB long, well organised into 117 sections with 39 code examples: a thorough specification that gives an agent plenty to work with.
With 3,433 GitHub stars, it is one of the more widely adopted skills in the catalogue.
Maintenance, license and trust
- The repository was last updated about 3 months ago, so hive-mind-advanced is actively maintained.
- Our last check on 2026-10-24 found the source still online.
- It is released under the Apache-2.0 license, a permissive license that allows use, modification and commercial use with attribution.
- Its trust signals score 100/100, with no cautions. These come from repository metadata, not a code audit — read the skill file before letting an agent act on it.
Safety scan
No issues foundOur scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful.
AI review by kimi-k2.7-code on 2026-10-02. Automated pattern scan on 2026-10-02. It catches known dangerous patterns, not every risk — read a skill before letting an agent act on it.
hive-mind-advanced compared with similar skills
All 4 of these similar skills score higher than hive-mind-advanced; compare them before choosing.
| Skill | Score | Stars | Updated | Format |
|---|---|---|---|---|
| hive-mind-advanced (this skill)by foryourhealth111-pixel | 95 | 3.4k | 3mo ago | SKILL.md |
| taste-skillby Leonxlnx | 100 | 116.8k | 8d ago | CLAUDE.md |
| ai-job-searchby ai-job-search-dev | 100 | 72.1k | today | CLAUDE.md |
| claude-howtoby luongnv89 | 100 | 47.3k | 2d ago | CLAUDE.md |
| lobehubby lobehub | 100 | 85.6k | today | MCP Server |
Frequently asked questions
- How do I install hive-mind-advanced?
- Run
npx skills add foryourhealth111-pixel/Vibe-Skills --skill hive-mind-advanced. The install tabs above show the steps for each supported agent. - Which AI agents does hive-mind-advanced work with?
- It is written for Claude Code, as a SKILL.md file. Other agents that read the same format can often use it too.
- Is hive-mind-advanced safe to use?
- Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful. It is Apache-2.0-licensed and scores 100/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
- Is hive-mind-advanced still maintained?
- The repository was last updated about 3 months ago, so hive-mind-advanced is actively maintained.
Skill content
View source on GitHubname: hive-mind-advanced description: Advanced Hive Mind collective intelligence system for queen-led multi-agent coordination with consensus mechanisms and persistent memory version: 1.0.0 category: coordination tags: [hive-mind, swarm, queen-worker, consensus, collective-intelligence, multi-agent, coordination] author: Claude Flow Team
Hive Mind Advanced Skill
Master the advanced Hive Mind collective intelligence system for sophisticated multi-agent coordination using queen-led architecture, Byzantine consensus, and collective memory.
Overview
The Hive Mind system represents the pinnacle of multi-agent coordination in Claude Flow, implementing a queen-led hierarchical architecture where a strategic queen coordinator directs specialized worker agents through collective decision-making and shared memory.
Core Concepts
Architecture Patterns
Queen-Led Coordination
- Strategic queen agents orchestrate high-level objectives
- Tactical queens manage mid-level execution
- Adaptive queens dynamically adjust strategies based on performance
Worker Specialization
- Researcher agents: Analysis and investigation
- Coder agents: Implementation and development
- Analyst agents: Data processing and metrics
- Tester agents: Quality assurance and validation
- Architect agents: System design and planning
- Reviewer agents: Code review and improvement
- Optimizer agents: Performance enhancement
- Documenter agents: Documentation generation
Collective Memory System
- Shared knowledge base across all agents
- LRU cache with memory pressure handling
- SQLite persistence with WAL mode
- Memory consolidation and association
- Access pattern tracking and optimization
Consensus Mechanisms
Majority Consensus Simple voting where the option with most votes wins.
Weighted Consensus Queen vote counts as 3x weight, providing strategic guidance.
Byzantine Fault Tolerance Requires 2/3 majority for decision approval, ensuring robust consensus even with faulty agents.
Getting Started
1. Initialize Hive Mind
# Basic initialization
npx claude-flow hive-mind init
# Force reinitialize
npx claude-flow hive-mind init --force
# Custom configuration
npx claude-flow hive-mind init --config hive-config.json
2. Spawn a Swarm
# Basic spawn with objective
npx claude-flow hive-mind spawn "Build microservices architecture"
# Strategic queen type
npx claude-flow hive-mind spawn "Research AI patterns" --queen-type strategic
# Tactical queen with max workers
npx claude-flow hive-mind spawn "Implement API" --queen-type tactical --max-workers 12
# Adaptive queen with consensus
npx claude-flow hive-mind spawn "Optimize system" --queen-type adaptive --consensus byzantine
# Generate Claude Code commands
npx claude-flow hive-mind spawn "Build full-stack app" --claude
3. Monitor Status
# Check hive mind status
npx claude-flow hive-mind status
# Get detailed metrics
npx claude-flow hive-mind metrics
# Monitor collective memory
npx claude-flow hive-mind memory
Advanced Workflows
Session Management
Create and Manage Sessions
# List active sessions
npx claude-flow hive-mind sessions
# Pause a session
npx claude-flow hive-mind pause <session-id>
# Resume a paused session
npx claude-flow hive-mind resume <session-id>
# Stop a running session
npx claude-flow hive-mind stop <session-id>
Session Features
- Automatic checkpoint creation
- Progress tracking with completion percentages
- Parent-child process management
- Session logs with event tracking
- Export/import capabilities
Consensus Building
The Hive Mind builds consensus through structured voting:
// Programmatic consensus building
const decision = await hiveMind.buildConsensus(
'Architecture pattern selection',
['microservices', 'monolith', 'serverless']
);
// Result includes:
// - decision: Winning option
// - confidence: Vote percentage
// - votes: Individual agent votes
Consensus Algorithms
- Majority - Simple democratic voting
- Weighted - Queen has 3x voting power
- Byzantine - 2/3 supermajority required
Collective Memory
Storing Knowledge
// Store in collective memory
await memory.store('api-patterns', {
rest: { pros: [...], cons: [...] },
graphql: { pros: [...], cons: [...] }
}, 'knowledge', { confidence: 0.95 });
Memory Types
knowledge: Permanent insights (no TTL)context: Session context (1 hour TTL)task: Task-specific data (30 min TTL)result: Execution results (permanent, compressed)error: Error logs (24 hour TTL)metric: Performance metrics (1 hour TTL)consensus: Decision records (permanent)system: System configuration (permanent)
Searching and Retrieval
// Search memory by pattern
const results = await memory.search('api*', {
type: 'knowledge',
minConfidence: 0.8,
limit: 50
});
// Get related memories
const related = await memory.getRelated('api-patterns', 10);
// Build associations
await memory.associate('rest-api', 'authentication', 0.9);
Task Distribution
Automatic Worker Assignment
The system intelligently assigns tasks based on:
- Keyword matching with agent specialization
- Historical performance metrics
- Worker availability and load
- Task complexity analysis
// Create task (auto-assigned)
const task = await hiveMind.createTask(
'Implement user authentication',
priority: 8,
{ estimatedDuration: 30000 }
);
Auto-Scaling
// Configure auto-scaling
const config = {
autoScale: true,
maxWorkers: 12,
scaleUpThreshold: 2, // Pending tasks per idle worker
scaleDownThreshold: 2 // Idle workers above pending tasks
};
Integration Patterns
With Claude Code
Generate Claude Code spawn commands directly:
npx claude-flow hive-mind spawn "Build REST API" --claude
Output:
Task("Queen Coordinator", "Orchestrate REST API development...", "coordinator")
Task("Backend Developer", "Implement Express routes...", "backend-dev")
Task("Database Architect", "Design PostgreSQL schema...", "code-analyzer")
Task("Test Engineer", "Create Jest test suite...", "tester")
With SPARC Methodology
# Use hive mind for SPARC workflow
npx claude-flow sparc tdd "User authentication" --hive-mind
# Spawns:
# - Specification agent
# - Architecture agent
# - Coder agents
# - Tester agents
# - Reviewer agents
With GitHub Integration
# Repository analysis with hive mind
npx claude-flow hive-mind spawn "Analyze repo quality" --objective "owner/repo"
# PR review coordination
npx claude-flow hive-mind spawn "Review PR #123" --queen-type tactical
Performance Optimization
Memory Optimization
The collective memory system includes advanced optimizations:
LRU Cache
- Configurable cache size (default: 1000 entries)
- Memory pressure handling (default: 50MB)
- Automatic eviction of least-used entries
Database Optimization
- WAL (Write-Ahead Logging) mode
- 64MB cache size
- 256MB memory mapping
- Prepared statements for common queries
- Automatic ANALYZE and OPTIMIZE
Object Pooling
- Query result pooling
- Memory entry pooling
- Reduced garbage collection pressure
Performance Metrics
// Get performance insights
const insights = hiveMind.getPerformanceInsights();
// Includes:
// - asyncQueue utilization
// - Batch processing stats
// - Success rates
// - Average processing times
// - Memory efficiency
Task Execution
Parallel Processing
- Batch agent spawning (5 agents per batch)
- Concurrent task orchestration
- Async operation optimization
- Non-blocking task assignment
Benchmarks
- 10-20x faster batch spawning
- 2.8-4.4x speed improvement overall
- 32.3% token reduction
- 84.8% SWE-Bench solve rate
Configuration
Hive Mind Config
{
"objective": "Build microservices",
"name": "my-hive",
"queenType": "strategic", // strategic | tactical | adaptive
"maxWorkers": 8,
"consensusAlgorithm": "byzantine", // majority | weighted | byzantine
"autoScale": true,
"memorySize": 100, // MB
"taskTimeout": 60, // minutes
"encryption": false
}
Memory Config
{
"maxSize": 100, // MB
"compressionThreshold": 1024, // bytes
"gcInterval": 300000, // 5 minutes
"cacheSize": 1000,
"cacheMemoryMB": 50,
"enablePooling": true,
"enableAsyncOperations": true
}
Hooks Integration
Hive Mind integrates with Claude Flow hooks for automation:
Pre-Task Hooks
- Auto-assign agents by file type
- Validate objective complexity
- Optimize topology selection
- Cache search patterns
Post-Task Hooks
- Auto-format deliverables
- Train neural patterns
- Update collective memory
- Analyze performance bottlenecks
Session Hooks
- Generate session summaries
- Persist checkpoint data
- Track comprehensive metrics
- Restore execution context
Best Practices
1. Choose the Right Queen Type
Strategic Queens - For research, planning, and analysis
npx claude-flow hive-mind spawn "Research ML frameworks" --queen-type strategic
Tactical Queens - For implementation and execution
npx claude-flow hive-mind spawn "Build authentication" --queen-type tactical
Adaptive Queens - For optimization and dynamic tasks
npx claude-flow hive-mind spawn "Optimize performance" --queen-type adaptive
2. Leverage Consensus
Use consensus for critical decisions:
- Architecture pattern selection
- Technology stack choices
- Implementation approach
- Code review approval
- Release readiness
3. Utilize Collective Memory
Store Learnings
// After successful pattern implementation
await memory.store('auth-pattern', {
approach: 'JWT with refresh tokens',
pros: ['Stateless', 'Scalable'],
cons: ['Token size', 'Revocation complexity'],
implementation: {...}
}, 'knowledge', { confidence: 0.95 });
Build Associations
// Link related concepts
await memory.associate('jwt-auth', 'refresh-tokens', 0.9);
await memory.associate('jwt-auth', 'oauth2', 0.7);
4. Monitor Performance
# Regular status checks
npx claude-flow hive-mind status
# Track metrics
npx claude-flow hive-mind metrics
# Analyze memory usage
npx claude-flow hive-mind memory
5. Session Management
Checkpoint Frequently
// Create checkpoints at key milestones
await sessionManager.saveCheckpoint(
sessionId,
'api-routes-complete',
{ completedRoutes: [...], remaining: [...] }
);
Resume Sessions
# Resume from any previous state
npx claude-flow hive-mind resume <session-id>
Troubleshooting
Memory Issues
High Memory Usage
# Run garbage collection
npx claude-flow hive-mind memory --gc
# Optimize database
npx claude-flow hive-mind memory --optimize
# Export and clear
npx claude-flow hive-mind memory --export --clear
Low Cache Hit Rate
// Increase cache size in config
{
"cacheSize": 2000,
"cacheMemoryMB": 100
}
Performance Issues
Slow Task Assignment
// Enable worker type caching
// The system caches best worker matches for 5 minutes
// Automatic - no configuration needed
High Queue Utilization
// Increase async queue concurrency
{
"asyncQueueConcurrency": 20 // Default: min(maxWorkers * 2, 20)
}
Consensus Failures
No Consensus Reached (Byzantine)
# Switch to weighted consensus for more decisive results
npx claude-flow hive-mind spawn "..." --consensus weighted
# Or use simple majority
npx claude-flow hive-mind spawn "..." --consensus majority
Advanced Topics
Custom Worker Types
Define specialized workers in .claude/agents/:
name: security-auditor
type: specialist
capabilities:
- vulnerability-scanning
Truncated for display — read the full file on GitHub.
Related Skills
taste-skill
116.8kAgent skills that give coding agents taste: art direction, reference images and a browser check instead of generic AI pages
ai-job-search
72.1kThe job search that runs on your machine. AI job application framework built on Claude Code: evaluate postings, tailor CVs, write cover letters, prep interviews. Fork it and own it.
claude-howto
47.3kA visual, example-driven guide to Claude Code — from basic concepts to advanced agents, with copy-paste templates that bring immediate value.
lobehub
85.6k🤯 LobeHub - Super-Agents on Your Team. Add super-agents to the chats your team already uses.They take on work by themselves and get better every week.
Languages
Trust signals
From repository metadata: license, adoption, age and documentation. Not a code audit — see the Safety scan above for what the skill file itself contains.