boot_installer

Install, update, repair, or health-check the openclaw environment. Use when the user says install openclaw, run the bootstrapper, update packages, fix a broken install, or check system health.

3,880 stars

Best use case

boot_installer is best used when you need a repeatable AI agent workflow instead of a one-off prompt.

Install, update, repair, or health-check the openclaw environment. Use when the user says install openclaw, run the bootstrapper, update packages, fix a broken install, or check system health.

Teams using boot_installer should expect a more consistent output, faster repeated execution, less prompt rewriting.

When to use this skill

  • You want a reusable workflow that can be run more than once with consistent structure.

When not to use this skill

  • You only need a quick one-off answer and do not need a reusable workflow.
  • You cannot install or maintain the underlying files, dependencies, or repository context.

Installation

Claude Code / Cursor / Codex

$curl -o ~/.claude/skills/boot-installer/SKILL.md --create-dirs "https://raw.githubusercontent.com/openclaw/skills/main/skills/alaminedits/boot-installer/SKILL.md"

Manual Installation

  1. Download SKILL.md from GitHub
  2. Place it in .claude/skills/boot-installer/SKILL.md inside your project
  3. Restart your AI agent — it will auto-discover the skill

How boot_installer Compares

Feature / Agentboot_installerStandard Approach
Platform SupportNot specifiedLimited / Varies
Context Awareness High Baseline
Installation ComplexityUnknownN/A

Frequently Asked Questions

What does this skill do?

Install, update, repair, or health-check the openclaw environment. Use when the user says install openclaw, run the bootstrapper, update packages, fix a broken install, or check system health.

Where can I find the source code?

You can find the source code on GitHub using the link provided at the top of the page.

Related Guides

SKILL.md Source

# boot_installer

Runs `{baseDir}/boot.sh` to manage the full openclaw environment lifecycle.
Requires Linux and root/sudo access. The script handles privilege escalation automatically.

## Slash commands

| Command | What it does |
|---|---|
| `/boot-install` | Fresh install of all components |
| `/boot-update` | Upgrade all packages to latest versions |
| `/boot-repair` | Deep clean and rebuild broken state |
| `/boot-check` | Health-check — verify all components |

## When to run which mode

- User says "install openclaw", "set it up", "run the installer" → **install** (no flag)
- User says "update", "upgrade packages" → **update**
- User says "repair", "fix", "something is broken", "rebuild" → **repair**
- User says "check", "status", "is everything installed", "health check" → **check**

## How to invoke

Always run as the calling user (sudo is handled internally by the script):

```bash
bash {baseDir}/boot.sh                # install
bash {baseDir}/boot.sh --update       # update
bash {baseDir}/boot.sh --repair       # repair
bash {baseDir}/boot.sh --check        # check — exits with number of failed checks (max 125)
```

Use the `exec` tool. Stream output so the user sees the spinner progress live.

## What gets installed (install mode)

- Node.js v24+ via NodeSource
- UV Python package manager (`~/.local/bin/uv`)
- Python 3.10 venv at `~/venv` with `scrapling[fetchers]`
- Chromium browser via Playwright (`~/.cache/ms-playwright`)
- NPM globals: `9router`, `openclaw@latest`, `clawhub`, `paperclipai`, `@presto-ai/google-workspace-mcp`, `mcporter`
- System symlinks in `/usr/local/bin`
- `9router` autostart via systemd user service (falls back to `.bashrc` hook)
- `openclaw-gateway-watchdog` systemd user service
- Mcporter Google Workspace connector

## Key paths

| Path | Purpose |
|---|---|
| `~/venv` | Python virtual environment |
| `~/.local/bin` | UV binary |
| `~/.local/npm/bin` | NPM global binaries |
| `~/.cache/ms-playwright` | Chromium browser |
| `~/.openclaw/workspace/skills` | Workspace skills |
| `~/.config/systemd/user/9router.service` | 9router systemd unit |
| `~/.config/systemd/user/openclaw-gateway-watchdog.service` | Gateway watchdog |

## After install

If `.bashrc` was modified, remind the user to reload their shell:
```bash
source ~/.bashrc
```

## Logs and errors

The script writes a full log to `/tmp/boot-YYYYMMDD-HHMMSS.log`. If a step fails, the last 15 lines of output are printed inline and the log path is shown. Surface that path to the user on failure.

## Check mode exit codes

`--check` exits with the count of failed checks (0 = all pass, 1–125 = N issues, capped at 125). Parse the exit code and tell the user how many checks failed and to run `bash {baseDir}/boot.sh` to repair.

## Troubleshooting

- **APT lock held** — script waits up to 120 s and kills blocking processes automatically. No user action needed.
- **Node.js wrong version** — script upgrades it automatically if the installed major is below 24.
- **Broken Python venv** — run `--repair` first, then run without flags to reinstall packages.
- **9router not starting** — logs at `~/.local/share/9router.log`. Falls back to `.bashrc` login hook if systemd unavailable.
- **Gateway watchdog not starting** — run `journalctl --user -u openclaw-gateway-watchdog -n 50`.
- **Permission errors on files** — run without flags (install mode); `_fix_ownership` runs at the end of every install/update.

Related Skills

safe-bootstrapper

3891
from openclaw/skills

Deterministic setup and remediation helper for installed OpenClaw skills. Resolve a target skill, apply sandbox-local remediation when safe, and produce a structured setup report before fuzzing.

boot-resume

3891
from openclaw/skills

Zero-cooperation session recovery after gateway restart. No checkpoints, no hooks, no agent involvement — just reads the evidence and picks up where it left off. Use when: the gateway was killed mid-task (SIGTERM, OOM, SIGKILL, crash), sessions were interrupted mid-turn with tool calls in progress, the agent stopped responding after a restart, a user reports the agent went silent after a crash, you need to manually check whether any sessions need recovery, or you want automatic resume without writing any checkpoint logic.

linux-installer

3891
from openclaw/skills

Installs, launches, and uninstalls Linux desktop apps by resolving the safest supported source first, then running a local helper CLI. Use when the user asks to install software like GIMP, Notepad++, or other desktop apps on Linux and wants the install command plus the command to launch or remove it.

gate-mcp-installer

3891
from openclaw/skills

One-click installer and configurator for Gate MCP (mcporter) in OpenClaw. Use when the user wants to (1) Install mcporter CLI tool, (2) Configure Gate MCP server connection, (3) Verify Gate MCP setup, or (4) Troubleshoot Gate MCP connectivity issues.

clawhub-quarantine-installer

3891
from openclaw/skills

Instala e audita skills do ClawHub em um ambiente de quarentena isolado para análise de segurança, permitindo revisar riscos antes de promover para produção. Use esta skill para testar habilidades de terceiros que o ClawHub sinaliza como suspeitas, investigar suas dependências e comportamento, e gerar relatórios de auditoria básicos.

openclaw-workspace-governance-installer

3891
from openclaw/skills

Install OpenClaw WORKSPACE_GOVERNANCE in minutes. Get guided setup, upgrade checks, migration, and audit for long-running workspaces.

zeroboot-vm-sandbox

3835
from openclaw/skills

Sub-millisecond VM sandboxes for AI agents using copy-on-write KVM forking via Zeroboot

---

3891
from openclaw/skills

name: article-factory-wechat

Content & Documentation

humanizer

3891
from openclaw/skills

Remove signs of AI-generated writing from text. Use when editing or reviewing text to make it sound more natural and human-written. Based on Wikipedia's comprehensive "Signs of AI writing" guide. Detects and fixes patterns including: inflated symbolism, promotional language, superficial -ing analyses, vague attributions, em dash overuse, rule of three, AI vocabulary words, negative parallelisms, and excessive conjunctive phrases.

Content & Documentation

find-skills

3891
from openclaw/skills

Helps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express interest in extending capabilities. This skill should be used when the user is looking for functionality that might exist as an installable skill.

General Utilities

tavily-search

3891
from openclaw/skills

Use Tavily API for real-time web search and content extraction. Use when: user needs real-time web search results, research, or current information from the web. Requires Tavily API key.

Data & Research

baidu-search

3891
from openclaw/skills

Search the web using Baidu AI Search Engine (BDSE). Use for live information, documentation, or research topics.

Data & Research