COMMUNITY-BUILT SPECIALISTS

Plug and Play AI Agents
for Any Domain

Zero Setup • Self-Extending • Open Source

Run octomind run doctor:blood and get a fully configured medical AI in 5 seconds.
The community builds the specialists. You just run them.
No config files. No MCP setup. No vendor lock-in.

Apache 2.0 7 AI Providers Dynamic MCP Adaptive Compression

Why Octomind?

The AI agent platform that actually works out of the box

Every AI Tool

  • 45 minutes of setup — MCP, prompts, configs
  • Rate limits mid-task — restart, lose context
  • Context rot — agent forgets hour-old decisions
  • One-size-fits-all — same tool for every task
  • Coding only — useless for doctors, lawyers, analysts

Octomind

  • Zero config — octomind run doctor:blood
  • Switch providers mid-session — no restart
  • Adaptive compression — 4-hour sessions work
  • Tap registry — specialists for any domain
  • Dynamic MCP — agents extend themselves

One Command. Any Specialist.

Run a specialist agent — no config, no setup, no MCP servers to manage

You: octomind run doctor:blood
Octomind: Installed medical specialist. Paste your lab results...
You: "Interpret these blood test results for a 45-year-old male"
AI: WBC slightly elevated — likely infection. LDL/HDL ratio normal...
You: octomind run devops:kubernetes
AI: Installed Kubernetes specialist with kubectl, helm, kustomize...
You: "Why is my pod stuck in CrashLoopBackOff?"
AI: Checking logs... Found OOMKilled. Memory limit too low...

Community builds specialists. You just run them.

octomind run developer:rust octomind run lawyer:contracts octomind run finance:analyst octomind run security:owasp
⟨/⟩

Tap Registry

Community-built specialists. One command: octomind run doctor:blood. All pre-configured.

Dynamic MCP

Agents register MCP servers mid-session. No restart. The agent extends itself.

Adaptive Compression

4-hour sessions stay coherent. Cache-aware cost analysis.

13 Providers

Zero Config

Install and run. Or customize with TOML. Your choice.

Rust Binary

Single binary. macOS + Linux. No dependencies.

Built-in Tools

Everything you need for development, research, and automation:

mcp

Register, enable, and disable MCP servers at runtime. Agents extend their own capabilities mid-session.

agent

Spawn specialized sub-agents on demand. Route tasks to the right AI for the job.

plan

Structured multi-step task tracking. Complex implementations with checkpoints.

view / text_editor

Smart file reading and editing. Batch operations with atomic multi-line edits.

ast_grep

AST-based code search and refactoring. Structural understanding, not regex.

shell / workdir

Full command execution. Background processes. Working directory management.

Session Commands

Switch providers, manage context, track costs — all mid-session:

⧉ Provider Switching

/model deepseek:v3

Switch providers mid-session. No restart. No lost context.

/model

View current model and available providers

⩔ Cost & Monitoring

/report

Detailed usage report with cost breakdown

/info

Real-time token usage and costs

⟨/⟩ Tap & Agents

octomind run <agent>

Run any specialist from the tap registry

octomind tap <repo>

Add a community or team tap

💡 Adaptive Compression

Work for 4 hours. The agent still knows what it decided in hour one. Cache-aware cost analysis. Pressure-level system. Automatic context management.

Zero Config. Infinitely Configurable.

Install and run — no config needed. Power users: full control when you need it.

Per-Role Models

Different models for different tasks. Cheap models for simple work, powerful models for complex reasoning.

Multi-Step Workflows

Each step can use a different model and toolset. Gather with Gemini Flash, critique with Claude Opus.

Spending Limits

Per-request and per-session limits. Never get surprised by a runaway agent.

Sandbox Mode

Landlock (Linux) / Seatbelt (macOS). Lock writes to working directory. Safe execution.

Pure TOML

No code required. All defaults in config-templates/default.toml. Override anything.

Dynamic MCP

Agents register new MCP servers at runtime. No restart. No config edit.

13 Providers. Zero Lock-in.

Hit a rate limit? Switch mid-session with /model. No restart. No lost context.

Provider Format Features
OpenRouter openrouter:provider/model Multi-provider access, 100+ models, caching, vision
OpenAI openai:model-name Direct API, GPT-4o, O1, vision, tools
Anthropic anthropic:model-name Claude 4, vision, tools, caching
DeepSeek deepseek:model-name Cost-effective, reasoning models, V3
Moonshot AI moonshot:model-name Kimi K2 series, vision, tools
MiniMax minimax:model-name Anthropic-compatible, thinking, tools
Z.ai z:model-name GLM models, cost-effective
Google google:model-name Vertex AI, Gemini 2.0, vision
Amazon amazon:model-name Bedrock, Claude on AWS
Cloudflare cloudflare:model-name Edge AI, fast inference, Llama
Ollama ollama:model-name Local models, Llama, Mistral, Qwen
Local local:model-name Any OpenAI-compatible endpoint
Cerebras cerebras:model-name Fastest inference, Llama variants

Built for Real Work

⟨/⟩

Tap Registry

Community-built specialists. One command: octomind run doctor:blood. The right model, tools, prompt — all pre-configured.

Dynamic MCP

Agents register new MCP servers at runtime. No restart. No config edit. Self-extending capabilities.

Adaptive Compression

Work for 4 hours. Agent still knows what it decided in hour one. Cache-aware cost analysis.

Provider Freedom

7 providers. Switch mid-session. Hit a rate limit? Type /model deepseek:v3 and keep going.

Zero Config

Install and run. No setup required. Power users: per-role models, workflows, spending limits, sandbox mode.

Sandbox Mode

Landlock (Linux) / Seatbelt (macOS). Lock writes to working directory. Safe execution.

Quick Start

1

Install Octomind

macOS and Linux. Single binary, no dependencies.

curl -fsSL https://raw.githubusercontent.com/muvon/octomind/master/install.sh | bash
2

Set API Key

OpenRouter gives you access to all 7 providers with one key.

export OPENROUTER_API_KEY="your_key"
3

Run a Specialist

Zero config. The tap registry handles everything.

octomind run doctor:blood

Or start a general session:

# Start an interactive session
octomind session

# Build from source
cargo install --git https://github.com/muvon/octomind.git

Open Source & Community

Apache 2.0 License

Free and open source software. Use, modify, and distribute freely.

Built with Rust

Performance, safety, and reliability. Single binary, macOS + Linux.

Community-Built Specialists

Contribute a tap agent — a single TOML file. Domain expertise welcome.