nanocoai/nanoclaw

Add Dashboard installs a local monitoring dashboard for NanoClaw, showing agent groups, sessions, channels, users, token usage, context windows, and real-time logs. Use it when you need observability into a NanoClaw deployment.

Add DeltaChat integrates DeltaChat as a messaging channel into NanoClaw by driving the @deltachat/stdio-rpc-server JSON-RPC subprocess directly. Use it to let users message the agent over encrypted email.

Add Discord Channel installs and registers the Discord bot adapter from the channels branch, adds the registration import, installs the pinned @chat-adapter/discord package, and validates with a build and test. It then stores Discord credentials, restarts the service, and resolves your DM channel so the agent can message you.

Add Emacs wires an Emacs editor into NanoClaw through a local HTTP bridge, opening a chat buffer and org-mode integration for asking questions, code review, and drafting from inside Emacs. Use it when you want to talk to the agent without leaving your editor.

Add Karpathy LLM Wiki sets up a persistent wiki knowledge base in NanoClaw using Karpathy's LLM Wiki pattern. Use it when a user asks to add a wiki, knowledge base, or llm wiki, and the agent needs to create sources, index, and a maintenance skill.

Add Linear Channel adds Linear support via the Chat SDK bridge so the agent participates in issue comment threads. It copies the Linear adapter, registers it, installs the package, and configures credentials and webhook wiring for team-routed comment handling.

Add Matrix Channel integrates Matrix messaging into NanoClaw via the Chat SDK by copying the adapter, registering it, installing dependencies, and patching ESM imports. Use it when the agent should send and receive messages on a Matrix homeserver.

Add Mnemon installs persistent graph-based memory via mnemon, so agents recall relevant past context before responding and remember new insights after each turn. Use it for Claude Code agents to persist memory across container restarts.

Add Ollama Integration adds a stdio-based MCP server that exposes local Ollama models as tools for the container agent, allowing it to offload work to local models and optionally manage the model library.

Add Ollama Provider routes an agent group to a local Ollama model instead of the Anthropic API, useful for running agents locally, cutting API costs, or experimenting with open-weight models.

Add OpenCode integrates OpenCode as an agent provider for NanoClaw by copying the provider files, registering the provider, installing dependencies, and rebuilding the container image. Use it to switch agent groups to OpenCode instead of the default Anthropic Agent SDK.

Add RTK installs the rtk token-compression proxy into agent containers and adds a PreToolUse hook so Bash calls are automatically routed through rtk for 60–90% token savings on dev commands.

Add Signal adds a native Signal channel to NanoClaw via signal-cli device-link, so an agent sends and receives messages as your existing phone number. Use it when wiring Signal DMs or groups to an agent group.

Add Slack Channel integrates Slack messaging into NanoClaw via the Chat SDK bridge. It copies the Slack adapter, registers it, installs the pinned package, builds and tests the integration, and configures Socket Mode or webhook credentials so agents can send and receive Slack messages.

Add Tavily Tool installs Tavily Search and Extract as keyless remote MCP tools for selected NanoClaw agent groups. Use it when you need web search or URL extraction without an API key.

Add Telegram copies the Telegram adapter, pairing helpers, and tests from the channels branch, registers them, and installs the pinned chat adapter package. Use it to add Telegram bot support to NanoClaw, including pairing a chat and wiring credentials.

Add iMessage connects an iMessage channel to NanoClaw using either a local macOS backend reading chat.db or a hosted backend via photon.codes. Use it when you want the agent to send and receive iMessages on any OS.

Atomic Chat Tool adds a stdio-based MCP server that exposes models running in the local Atomic Chat desktop app as tools for the container agent. Use it to let Claude offload work to local models served by Atomic Chat on its OpenAI-compatible API.

Channel Integration adds GitHub support via the Chat SDK bridge so the agent can participate in PR and issue comment threads.

Clidash adds a zero-dependency, read-only web dashboard that derives its tabs and tables at runtime from any CLI that lists resources as JSON. Use it to monitor NanoClaw agent groups, sessions, channels, users, roles, message activity, logs, and group files.

Codex Provider Setup installs OpenAI Codex as an agent provider for NanoClaw, enabling planning, tool orchestration, MCP tools, server-side history, and session resume alongside or instead of Claude. Use it to switch a group's provider to Codex via `ncl groups config update --provider codex`.

Google Chat Channel Integration adds Google Chat support to NanoClaw via its Chat SDK bridge. It copies the adapter, registers it, installs the pinned package, and runs build and integration tests. Use it when setting up a Google Chat channel for an agent group.

Microsoft Teams Integration adds Microsoft Teams support via the Chat SDK bridge so an agent can chat in team channels, group chats, and direct messages. Use it when you need to register a Teams bot, provision an Entra app, and connect the webhook endpoint.

Resend Email Integration adds a Resend email channel to NanoClaw by copying the adapter, registering the import, installing the package, and running build and tests. Use it to let an AI agent send and receive email asynchronously.

macOS Status Bar Indicator adds a persistent macOS menu bar icon for NanoClaw that shows running status and provides Start, Stop, and Restart controls. Use it to manage the NanoClaw service on macOS.