> ## Documentation Index
> Fetch the complete documentation index at: https://mintlify.com/nyldn/claude-octopus/llms.txt
> Use this file to discover all available pages before exploring further.
# Introduction
> Claude Octopus is a multi-AI orchestration framework for Claude Code that coordinates three AI providers with distinct roles, adversarial review, and consensus gates.
# Introduction
Claude Octopus is a Claude Code plugin that turns one model into three. It orchestrates Codex, Gemini, and Claude with distinct roles, adversarial review, and consensus gates — so no single model's blind spots slip through.
## What problem does it solve?
Other multi-AI tools run providers in parallel and hand you three answers. Claude Octopus assigns each model a distinct role — Codex for implementation depth, Gemini for ecosystem breadth, Claude for synthesis — then enforces a 75% consensus gate before anything ships. Disagreements get caught, not ignored.
## Key features
### Multi-AI orchestration
Coordinate three AI providers across every workflow:
| Provider | Role |
| ---------- | ---------------------------------------------------------------------- |
| **Codex** | Implementation depth — code patterns, technical analysis, architecture |
| **Gemini** | Ecosystem breadth — alternatives, security review, research synthesis |
| **Claude** | Orchestration — quality gates, consensus building, final synthesis |
Providers run in parallel for research, sequentially for problem scoping, and adversarially for review. A 75% consensus quality gate prevents questionable work from shipping.
### Double Diamond methodology
Every task moves through four structured phases adapted from the UK Design Council's framework:
| Phase | Command | What happens |
| ------------ | ---------------- | ----------------------------------------- |
| **Discover** | `/octo:discover` | Multi-AI research and broad exploration |
| **Define** | `/octo:define` | Requirements clarification with consensus |
| **Develop** | `/octo:develop` | Implementation with quality gates |
| **Deliver** | `/octo:deliver` | Adversarial review and go/no-go scoring |
Run phases individually or all four with `/octo:embrace`. Configure autonomy: supervised (approve each phase), semi-autonomous (intervene on failures), or autonomous (run all four).
### 33 specialized personas
Not generic agents. Specialized personas that activate automatically based on your request:
* **Software Engineering (11)**: backend-architect, frontend-developer, tdd-orchestrator, debugger, security-auditor, and more
* **Specialized Development (6)**: python-pro, typescript-pro, database-architect, ai-engineer, cloud-architect, graphql-architect
* **Documentation & Communication (5)**: docs-architect, product-writer, exec-communicator, academic-writer, mermaid-expert
* **Research & Strategy (4)**: research-synthesizer, strategy-analyst, ux-researcher, business-analyst
* **Business & Compliance (3)**: finance-analyst, legal-compliance-advisor, marketing-strategist
* **Creative & Design (4)**: Specialized visual and content creation
When you say "audit my API for vulnerabilities," security-auditor activates. When you say "write a research paper," academic-writer takes over.
### 46 commands
Eight commands that orchestrate up to three AI providers, apply quality gates, and produce deliverables:
```bash theme={null}
/octo:embrace build stripe integration # Full lifecycle: research → define → develop → deliver
/octo:factory "build a CLI" # Autonomous pipeline — spec in, software out
/octo:debate monorepo vs microservices # Structured three-way AI debate with consensus
/octo:research htmx vs react in 2026 # Multi-source synthesis from three AI providers
/octo:review # Code review with security + 4x10 scoring
/octo:tdd create user auth # Red-green-refactor with test discipline
/octo:security # OWASP vulnerability scan + remediation
/octo:prd mobile checkout redesign # AI-optimized PRD with 100-point scoring
```
Including: debug, extract, deck, docs, schedule, parallel, sentinel, brainstorm, claw, doctor, and more.
**Don't remember the command name?** Just describe what you need:
```
/octo research microservices patterns → routes to discover phase
/octo build user authentication → routes to develop phase
/octo compare Redis vs DynamoDB → routes to debate
```
The smart router parses your intent and selects the right workflow.
## How it works at a high level
### Three brains, one workflow
1. **Parallel research** — Multiple AI providers research different angles simultaneously
2. **Consensus building** — 75% agreement required before advancing
3. **Quality gates** — Each phase validates before the next begins
4. **Adversarial review** — Providers challenge each other's work ("the squeeze")
### Dark Factory mode
Takes a spec and autonomously runs the full pipeline — research, define, develop, deliver — with holdout testing and satisfaction scoring. You review the output, not every step.
```bash theme={null}
/octo:factory "build a CLI that converts CSV to JSON"
```
### Works with just Claude. Scales to three.
Zero external providers needed to start. You get every persona, every workflow, every skill on day one. Add Codex or Gemini and multi-AI orchestration lights up — parallel research, adversarial debate, cross-model review.
### Full cost transparency
Visual indicators show exactly which providers are active and whose billing they hit:
| Indicator | Meaning | Cost Source |
| --------- | -------------------- | ------------------------- |
| 🐙 | Multi-AI mode active | Multiple APIs |
| 🔴 | Codex CLI executing | User's OPENAI\_API\_KEY |
| 🟡 | Gemini CLI executing | User's GEMINI\_API\_KEY |
| 🔵 | Claude subagent | Included with Claude Code |
Smart cost routing picks the cheapest provider that can handle the job. No hidden API calls, no surprise charges.
## Who should use it?
Claude Octopus is built for:
Building features that need research, design, implementation, and validation in one workflow
Projects requiring multiple perspectives and adversarial review to catch edge cases
Anyone who wants structured quality gates and multi-AI validation without manual coordination
Knowledge workers who need deep research synthesis and consensus-driven insights
**Start simple:** Use Claude Octopus with just Claude built-in. Add external providers (Codex, Gemini) when you need multi-AI features like parallel research and adversarial debate.
## What's next?
Get to your first workflow in under 5 minutes
Detailed setup with provider configuration