Octomind Documentation: Open-Source AI Coding Agent CLI
Octomind is an open-source AI coding agent and agent runtime for the terminal: one Rust binary, any model, MCP-native.
Octomind is an open-source AI coding agent and agent runtime for the terminal: one Rust binary, any model, MCP-native. Use these guides to install, configure, and operate Octomind's CLI agent runtime, or to contribute to its Rust code. Start with a terminal session, then follow the guides for automation, integrations, and development.
Start Here
Install the binary and sign in to an Octomind account to use the shipped OctoHub model default:
curl -fsSL https://octomind.run/install.sh | bash
octomind login
octomindoctomind login asks you to confirm a code in the browser and CLI, then stores the hub key in the user-scope
config/.env and account credentials in config/auth.json. It does not change your model selection. New configs use
assistant:concierge as the default tag and octohub:auto as the model name. See
Installation and Quickstart for the complete setup.
Configure
Inspect and validate your configuration before adding roles or tools:
octomind config --show
octomind config --validateThe default config is under ~/.local/share/octomind/config/ on Linux/macOS and %LOCALAPPDATA%/octomind/config/ on
Windows. OCTOMIND_DATA_DIR relocates Octomind state; for example:
OCTOMIND_DATA_DIR="$PWD/.octomind-data" octomind config --showUse Configuration for file merging and overrides, or AI Providers to supply provider credentials yourself.
Operate
Create or resume a named interactive session:
octomind run --name project-review
# After exiting, continue the same session:
octomind run --resume project-reviewInside a session, discover commands and inspect active tools:
/help
/mcp
/exitFor automation, pipe the prompt through stdin and choose plain or jsonl output:
printf '%s\n' 'Summarize the responsibilities of an MCP server in three sentences.' | octomind run --format plainrun takes a role/tag positional argument, not a message argument. See Daemon &
Hooks for persistent workers and Workflows for multi-step
runs.
Usage Guide
| Document | Description |
|---|---|
| Installation | Recommended setup, alternative installs, and shell completions |
| Quickstart | First session, common commands, and non-interactive use |
| Configuration | Config locations, merging, models, roles, and MCP servers |
| Providers | OctoHub, provider credentials, model selection, and local CLI backends |
| Sessions | Interactive sessions, persistence, and multimodal input |
| Roles | Roles, prompts, permissions, and tool access |
| MCP Tools | Built-in tools and runtime tool management |
| Compression | Automatic context compression |
| Workflows | Multi-step AI processing workflows |
| Commands & Layers | Custom commands, layers, agents, and prompts |
| Structured Output | JSON Schema output for automation |
| Editor Integration | Neovim, Zed, and JetBrains setup |
| Learning | Cross-session adaptive learning |
| Supervisor | Completion checks, planning, condensation, and learning control |
| Skills | Auto-activating skills and validators |
| Token Efficiency | Context and capability efficiency |
| Local Tools | Project-local scripts exposed as MCP tools |
| Guardrails | Deterministic project policies and hooks |
Integration Guide
| Document | Description |
|---|---|
| WebSocket Server | Remote sessions over WebSocket |
| ACP Protocol | Agent Client Protocol integration |
| Daemon & Hooks | Long-running sessions and webhook listeners |
| Tap System | Agent, skill, capability, and workflow registries |
Use Cases
| Document | Description |
|---|---|
| CI/CD Code Review | Automated review with structured output |
| Event-Driven Agent | Daemon sessions driven by webhooks |
| Custom Workflow | Multi-stage development workflows |
| Web Dashboard | Embedding sessions through WebSocket |
| Multi-Agent Delegation | Delegating work to specialized agents |
| Dynamic MCP Servers | Runtime tool-server configuration |
| Scheduled Tasks | Timed messages and recurring work |
| Long-Running Development | Named sessions and resume workflows |
| Custom Hooks | Script-backed webhook integration |
| Safe Agent: Sandbox & Guardrails | Block destructive commands and verify the agent’s work |
| Predictable AI Spend | Spending checkpoints, per-role models, and usage reports |
| Make “Done” Mean Done | Supervisor gate, plan checks, and a test validator |
| Project Conventions via Skills | Auto-activating skills and convention validators |
| Learn from Past Sessions | Carry corrections into the next session |
| Project Scripts as Tools | Let the agent run your checks through local tools |
| Build & Share a Specialist | Create, publish, and install a tap agent |
| Understand a Codebase | Source-backed onboarding to an unfamiliar repo |
| Private Code, Local Models | Route private roles to Ollama, others to the cloud |
| Second Opinion via Fan-out | Parallel branches and a judge step |
| Research with Dynamic Fan-out | Planner-driven parallel investigations |
| Implement–Review–Fix Loop | Bounded graph workflow until review passes |
| Release Notes from Git History | Plain drafts and schema-validated release data |
| Calibrated Checks | Probability-scored done, scope, and vagueness checks in scripts, validators, and pipes |
Development Guide
| Document | Description |
|---|---|
| Building from Source | Rust setup and development builds |
| Architecture | Source modules and internal flows |
| MCP Server Development | Building MCP servers for Octomind |
| Learning Benchmark | Retrieval and consolidation contract benchmark |
Common Questions
If the shell cannot find the binary after the default installation, add the installer directory to your current shell's path:
export PATH="$HOME/.local/bin:$PATH"
octomind --versionIf login cannot open a local browser, print the URL instead:
octomind login --no-browserIf a piped command hangs, ensure its producer closes stdin; Octomind reads the complete prompt before starting. For
configuration and credential errors, start with octomind config --validate above and the guides below.
Troubleshooting and Reference
| Document | Description |
|---|---|
| Common Issues | Installation, configuration, provider, and session problems |
| Migration Guide | Upgrading legacy configurations |
| CLI Reference | CLI subcommands and flags |
| Session Commands | Interactive slash commands |
| Config Reference | Configuration fields and defaults |
| Environment Variables | Credentials, overrides, and runtime variables |