row-bot

Name: row-bot
Author: siddsachar

Verified

Row-Bot - Personal AI Sovereignty. A local-first AI assistant with integrated tools, a personal knowledge graph, voice, vision, shell, browser automation, scheduled tasks, health tracking, and messaging channels. Run locally via Ollama or add opt-in cloud models. Your data stays on your machine.

1,603stars

182forks

Python

Installation

# Add to your Claude Code skills
git clone https://github.com/siddsachar/row-bot

Getting Started

Guides for using ai agents skills like row-bot.

Security ReportVerified

Last scanned: 6/5/2026

{
  "issues": [],
  "status": "PASSED",
  "scannedAt": "2026-06-05T08:07:33.723Z",
  "npmAuditRan": true,
  "pipAuditRan": false
}

README.md

Frequently Asked Questions

What is row-bot?

row-bot is an open-source ai agents skill for AI coding assistants such as Claude Code, Codex CLI, and ChatGPT, built by siddsachar. Row-Bot - Personal AI Sovereignty. A local-first AI assistant with integrated tools, a personal knowledge graph, voice, vision, shell, browser automation, scheduled tasks, health tracking, and messaging channels. Run locally via Ollama or add opt-in cloud models. Your data stays on your machine. It has 1,603 GitHub stars.

Is row-bot safe to use?

Yes. row-bot passed SkillsLLM's automated security scan — a dependency vulnerability audit plus prompt-injection heuristics — with no high-severity issues. You can read the full report in the Security Report section on this page.

How do I install row-bot?

Clone the repository with "git clone https://github.com/siddsachar/row-bot" and add it to your Claude Code skills directory (see the Installation section above).

What programming language is row-bot written in?

row-bot is primarily written in Python. It is open-source under siddsachar on GitHub, so you can review or fork the full source.

Are there alternatives to row-bot?

Yes. SkillsLLM lists many other AI Agents skills you can browse and compare side by side. Open the AI Agents category from the badge at the top of this page, or use the Related Skills and comparison links further down to weigh row-bot against similar tools.

Agentic AI for Beginners

Build your first AI agent from scratch - tool use, ReAct pattern, memory, deployment

41 minBeginner

Comments (0)

to leave a comment.

No comments yet. Be the first to share your thoughts!

Related Skills

ECC

by affaan-m

The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.

276,163

opencove spec-kitty

⚠️ Third-Party Software Notice

This skill is third-party open-source software developed and hosted independently on GitHub. SkillsLLM is an informational directory and does not control or maintain the underlying repository.

Any security checks, ratings, or warnings displayed by SkillsLLM are automated and limited in scope. They do not constitute a security certification or guarantee that the software is safe, error-free, or free from malicious code, vulnerabilities, compromised dependencies, or prompt-injection risks.

Review the source code, permissions, dependencies, and configuration before installing or running any third-party skill. Use is at your own risk. To the maximum extent permitted by applicable law, SkillsLLM is not liable for losses arising from third-party software.

Read the Terms of Service

Row-Bot is a local-first desktop AI assistant for doing real work with models, memory and tools. You talk to it in conversations; it remembers what matters in a local knowledge graph, works in the files, code folders, designs, workflows and channels you give it, delegates to focused agents, and asks before anything risky. It runs local models through Ollama, provider API keys, ChatGPT, Claude or Grok subscriptions, or any OpenAI-compatible endpoint, and keeps your data on your machine. One React app serves the desktop window, browsers, phones and server mode.

Features

Conversations that do the work - ask for a deck, an app or a report and Row-Bot creates the design or code folder and keeps working in it; setup cards turn on a missing tool or connect an account in place.
Goals and agents - goals that keep going until done and pause when they stop making progress; delegated agents with their own conversations, Stop and Message; reusable agent profiles.
Approvals wherever you are - risky actions wait for Approve or Deny in the conversation, on Home, in the attention indicator, in Buddy or on a connected channel.
Memory and knowledge - a personal knowledge graph with recall, review, Dream Cycle refinement, documents, and an Obsidian-compatible wiki vault.
Design and code panels - decks, documents and mockups with Present, Review, export (PDF, HTML, PNG, PPTX) and publish; code folders with changes, files, Git, checks, an interactive terminal and an optional Docker sandbox.
Workflows - scheduled, webhook and chained runs with approvals, delivery to channels and run history.
Tools and extensions - web search, files, shell, managed browser, opt-in Computer Use on Windows and macOS, Gmail and Calendar, image and video generation, skills, plugins, MCP servers and custom tools.
Channels and voice - Telegram, WhatsApp, Discord, Slack and SMS; dictation, Talk and read-aloud with local Whisper and Kokoro.
Monitor with fixes - health checks that run by themselves, with one fix per problem, and Insights that suggest improvements.
Devices and remote access - connect a phone or another computer with a QR code over Tailscale, your Wi-Fi or a public link; a Buddy desktop overlay on Windows and macOS; backup and restore of your profile.

Platforms

Platform	Package	Notes
Windows 10/11, 64-bit	Installer	Native window, tray, Buddy overlay, Computer Use.
macOS 12+, Apple Silicon and Intel	DMG	Native window, tray, Buddy overlay, Computer Use.
Linux x86_64 (glibc)	Tarball and one-line installer	Opens in your browser; a native window and tray need GTK or Qt and AppIndicator.
Docker, amd64 and arm64	`ghcr.io/siddsachar/row-bot`	Authenticated single-owner server.

Ollama is optional, for local models only. System requirements are in the installation guide.

Install

Download the Windows installer or the macOS DMG from GitHub Releases and run it. Upgrades and repairs keep your data.

On Linux:

curl -fsSL https://raw.githubusercontent.com/siddsachar/row-bot/main/installer/install-linux.sh | bash
# or a specific published release (replace X.Y.Z):
curl -fsSL https://raw.githubusercontent.com/siddsachar/row-bot/main/installer/install-linux.sh | bash -s -- X.Y.Z

The installer takes the latest release by default, verifies the tarball's SHA-256, installs under ~/.local/share/row-bot and creates ~/.local/bin/row-bot.

With Docker, from a checkout of this repository (pin X.Y.Z to a release with a published container):

export ROW_BOT_IMAGE=ghcr.io/siddsachar/row-bot:X.Y.Z
docker compose -f deploy/docker/compose.yaml up --detach
docker compose -f deploy/docker/compose.yaml exec row-bot \
  row-bot access invite --layout desktop --origin http://127.0.0.1:8080

Read Docker and VPS operations before exposing it beyond loopback.

Upgrading from 4.x? Read the v5.0.0 release notes first: the old NiceGUI interface is gone and several places moved. Users still on Thoth or an early Row-Bot build should first run a previous migration-capable Row-Bot release.

From source

You need Python 3.12 or 3.13, uv, Git, and Node.js 24.15+ with npm 11 to build the client.

git clone https://github.com/siddsachar/row-bot.git
cd row-bot
python -m pip install "uv>=0.7,<1.0"
uv sync --locked --all-extras --group test
cd frontend && npm ci --ignore-scripts --no-audit --no-fund && npm run build && cd ..
uv run python launcher.py

launcher.py starts the tray and opens the app at http://localhost:8080/app-v2/ (the next free port if 8080 is busy); on Linux it opens your browser. Starting it again while it runs shows the running app. Other entry points:

uv run python launcher.py serve --port 8080   # authenticated headless server
uv run python app.py                          # server only, no tray or window

Dependency changes go through pyproject.toml and uv.lock; never edit requirements.txt by hand (see AGENTS.md).

Quick start

On first launch Row-Bot asks How should Row-Bot think?: on this computer (Ollama), with your subscription (ChatGPT, Claude or Grok), with an API key, or with a custom endpoint. Your pick becomes the default model, a short test runs, and Home opens.
Start a conversation and ask for what you need, for example Create a six-slide pitch deck for my startup, Review this repo and suggest the highest-risk issues, Remember that my mom's birthday is March 15 or Remind me to call the dentist tomorrow at 9am.
Press Ctrl+K (⌘K on macOS) to find conversations, commands and settings by plain words, such as "connect telegram" or "dark mode".
Setup Center (Continue setup on Overview, or search for it with Ctrl+K) walks through the rest: knowledge, workflows, channels, accounts, tools and voice.

The first launch guide covers each path.

Configuration and data

Data folder: ~/.row-bot (%USERPROFILE%\.row-bot on Windows), or the folder in ROW_BOT_DATA_DIR. It holds conversations, memory, workflows, settings and logs (row_bot_app.log and logs/row_bot.log).
Workspace folder: ~/Documents/Row-Bot by default, where file tools, exports and backups go; change it in Settings › System.
Secrets: provider keys and sign-ins are stored in Windows Credential Manager, the macOS Keychain or Linux Secret Service/KWallet; Docker uses encrypted records with a separate key volume. JSON settings hold metadata only.
Backups: Settings › Data › Back up now makes a zip without secrets.
Reference: models and providers, environment and configuration, data storage and command-line options.

Remote access and server mode

Row-Bot is a single-owner, multi-device application: one owner can use the same instance from several trusted browsers, but it is not multi-user hosting. A device joins with a one-time invitation (a QR code or link that expires after 10 minutes) and then holds a revocable session. Connect devices from Settings › Devices & remote access, or from a trusted terminal:

Path	Use it when	Boundary
Tailscale Serve	Private access across your devices (recommended).	Row-Bot never installs Tailscale, signs in, enables Funnel or overwrites another app's route.
Direct LAN	A trusted network.	Plain HTTP is unencrypted; prefer Tailscale or HTTPS outside a trusted LAN.
SSH tunnel	A remote host with no published port.	Keep Row-Bot on loopback and forward a port.
Docker	A headless, isolated instance.	Compose publishes to host loopback by default.
HTTPS reverse proxy or VPS	You manage DNS, TLS and the proxy.	Set `ROW_BOT_PUBLIC_URL`, `ROW_BOT_ALLOWED_HOSTS` and `ROW_BOT_TRUSTED_PROXY_CIDRS` exactly; never buffer `/api/v1/events`.

row-bot serve --host 127.0.0.1 --port 8080
row-bot access invite --layout desktop --origin http://127.0.0.1:8080
row-bot access list
row-bot access doctor

If the owner loses every browser session, create a new invitation from a trusted local terminal, SSH session or docker compose exec. Back up the complete active ROW_BOT_DATA_DIR while Row-Bot is stopped (with Docker, the data and encryption-key volumes together). Details: Remote access and server mode and [Docker and V