claude-octopus
Surface AI blindspots before you ship. Put up to 8 AI models on every research, design or coding task.
Claude Octopus is a Claude Code plugin that surfaces AI blind spots by running up to 8 different AI models on the same task simultaneously. It integrates twelve external AI providers (including Codex, Copilot, Qwen, Ollama, Perplexity, OpenRouter, Grok, and others) alongside Claude, using a 75% consensus gate to flag disagreements before deployment. The plugin provides 54 specialized commands for research, design, coding, review, and security tasks, with 31 role-specific AI personas and structured Double Diamond workflows (Discover → Define → Develop → Deliver). It remains dormant until explicitly invoked via /octo:* commands, making it ideal for high-stakes decisions, adversarial code review, and multi-model validation without disrupting normal Claude Code usage.
Key Features
Use Cases
- 01Adversarial code review where multiple AI models cross-check implementation decisions and surface edge cases before production
- 02Security audits using /octo:security to run OWASP vulnerability scans across multiple models with consensus-based threat detection
- 03Architecture debates via /octo:council to get structured multi-model recommendations with quorum gates for high-stakes technology decisions
- 04Research tasks requiring /octo:research with breadth modes that fan out queries to multiple providers and return attributed sources
- 05Full feature development using /octo:embrace to run research → define → develop → deliver phases with quality gates between each step
- 06TDD workflows via /octo:tdd that enforce red-green-refactor discipline with multi-model test validation before implementation
Related Plugins
View moreandrej-karpathy-skills
A single CLAUDE.md file to improve Claude Code behavior, derived from Andrej Karpathy's observations on LLM coding pitfalls.
claude-mem
Persistent Context Across Sessions for Every Agent – Captures everything your agent does during sessions, compresses it with AI, and injects relevant context back into future sessions. Works with Claude Code, OpenClaw, Codex, Gemini, Hermes, Copilot, OpenCode + More
Understand-Anything
Graphs that teach > graphs that impress. Turn any code into an interactive knowledge graph you can explore, search, and ask questions about. Works with Claude Code, Codex, Cursor, Copilot, Gemini CLI, and more.
rtk
CLI proxy that reduces LLM token consumption by 60-90% on common dev commands. Single Rust binary, zero dependencies
claude-octopus — FAQ
What is Claude Octopus and what does it do?+
Claude Octopus is a Claude Code plugin that runs multiple AI models (up to 12 external providers plus Claude) on the same task to catch blind spots and disagreements. It provides 54 specialized commands for research, design, coding, review, and security workflows with a 75% consensus gate that flags low-agreement work before it ships.
How do I install the Claude Code plugin?+
Run 'claude plugin marketplace add https://github.com/nyldn/plugins.git' then 'claude plugin install octo@nyldn-plugins' in your terminal. Inside Claude Code, run /octo:setup to configure detected providers. Claude Code v2.1.14 or newer is required.
Do I need API keys or subscriptions for external providers?+
No external providers are required to start; Claude is built in. Five providers (Codex, Antigravity CLI, Copilot, Cursor CLI, Ollama) use existing subscriptions or run locally for free. Others like Perplexity, OpenRouter, and Qwen require API keys, which you add one at a time as needed.
Which AI clients does Claude Octopus work with?+
Claude Octopus works primarily with Claude Code (as a plugin) and Cursor IDE (as an MCP server). It also supports Codex CLI, OpenCode, and Factory AI (Droid) via their respective plugin systems.
Is Claude Octopus free to use?+
The plugin itself is free and open-source (MIT license). Cost comes from external AI provider usage: five providers cost nothing extra if you already have access (Codex via ChatGPT subscription, Copilot, Antigravity, Cursor CLI, local Ollama). API-based providers like Perplexity and OpenRouter bill per token; typical runs range from $0.01 for quick tasks to $1-6+ for full lifecycle workflows.
Does installing Claude Octopus change how Claude Code normally works?+
No. Claude Octopus is dormant by default and only activates when you explicitly run /octo:* commands. Ordinary prompts continue to use Claude normally unless you opt into the auto-router with OCTOPUS_AUTO_ROUTER_MODE=suggest or invoke.
How do I install claude-octopus?+
Open the source repository on GitHub and follow its README. claude-octopus is a plugin — MCP Agents Market links you directly to the official repo.
Is claude-octopus free?+
claude-octopus is an open-source project hosted on GitHub. Check the repository for its license and any usage requirements.