Multi-Agent System

Hubify Labs runs a hierarchical multi-agent system designed to mirror the structure of a real research group. The system is built on three principles: hierarchy for efficiency, cross-model review for accuracy, and full transparency for trust.

Architecture

                  ┌──────────────┐
                  │   Captain    │  ← You
                  │  (Human)     │
                  └──────┬───────┘
                         │
                  ┌──────▼───────┐
                  │ Orchestrator │  ← Claude Opus 4.8
                  │  (Router)    │
                  └──┬───┬───┬───┘
            ┌────────┘   │   └────────┐
     ┌──────▼──────┐ ┌──▼────┐ ┌─────▼──────┐
     │Research Lead│ │Paper  │ │Compute Lead│
     │ (Opus 4.8) │ │Lead   │ │ (Haiku 4.5)│
     └──┬────┬────┘ └──┬────┘ └─────┬──────┘
        │    │         │             │
     Workers Workers  Workers     Workers
     (Haiku) (Haiku)  (Haiku)     (Haiku)

Reasoning-Based Routing

Every task has a reasoning requirement. The orchestrator routes accordingly:

High Reasoning
Medium Reasoning
Low Reasoning

Models: Claude Opus 4.8, GPT-5.5, Gemini 3.1 Pro, Grok 4Tasks: Research strategy, paper drafting, peer review, novel scientific analysis, hypothesis generation, cross-survey interpretation.Handled by: Orchestrator or Lead agents.

Cross-Model Peer Review

Every significant output is reviewed by models from at least two different providers. Same-provider review is not permitted.

The review matrix uses five providers across five labs:

Output from	Reviewed by
Claude Opus 4.8	GPT-5.5, Gemini 3.1 Pro, Grok 4, Sonar Pro
GPT-5.5	Claude Opus 4.8, Gemini 3.1 Pro, Grok 4, Sonar Pro
Gemini 3.1 Pro	Claude Opus 4.8, GPT-5.5, Grok 4, Sonar Pro
Grok 4	Claude Opus 4.8, GPT-5.5, Gemini 3.1 Pro, Sonar Pro

Reviews check for:

Factual accuracy and hallucination detection
Logical consistency with prior results
Mathematical and statistical correctness
Missing citations or prior work
Overstatements and unsupported claims

Activity Feed

All agent communication is visible in the Activity Feed, a real-time, color-coded stream:

[10:42] [done] Research Lead completed EXP-054 (MCMC base chain)
[10:43] [info] Orchestrator -> Paper Lead: "Integrate EXP-054 results into Section 4"
[10:44] [info] Paper Lead -> Draft Worker: "Update posterior table with new chain means"
[10:45] [warn] QC Worker flagged EXP-055: convergence R-hat = 1.08 (threshold: 1.05)
[10:46] [crit] Research Lead escalated: "EXP-055 needs longer chains. Request H200 pod."

Tilldone Pattern

When a worker fails a task, the lead agent takes over rather than just reporting the failure:

Worker attempts the task
Worker fails (error, bad output, QC failure)
Lead agent receives the failure with full context
Lead agent executes the task itself using higher reasoning
If the lead also fails, it escalates to the orchestrator

This pattern ensures tasks complete without constant human intervention.

Adding Agents

# Add a specialized lead
hubify agent add --role lead --name "Cosmology Lead" --model claude-opus \
  --specialty "MCMC analysis, CMB power spectra, dark energy constraints"

# Add workers
hubify agent add --role worker --name "Figure Generator" --model claude-haiku
hubify agent add --role worker --name "Data Processor" --model claude-haiku

# View the full roster
hubify agent list --tree

Agent Metrics

Each agent tracks:

Tasks completed vs failed
Average task duration
QC pass rate
Review acceptance rate
Cost per task

Getting Started

Core Concepts

Features

Guides

Integrations

Resources

Multi-Agent System

Multi-Agent System

Architecture

Reasoning-Based Routing

Cross-Model Peer Review

Activity Feed

Tilldone Pattern

Adding Agents

Agent Metrics

​Multi-Agent System

​Architecture

​Reasoning-Based Routing

​Cross-Model Peer Review

​Activity Feed

​Tilldone Pattern

​Adding Agents

​Agent Metrics

Multi-Agent System

Architecture

Reasoning-Based Routing

Cross-Model Peer Review

Activity Feed

Tilldone Pattern

Adding Agents

Agent Metrics