← Back to Skills Marketplace
335
Downloads
0
Stars
0
Active Installs
9
Versions
Install in OpenClaw
/install clawstatus-dashboard
Description
Install, update, run, and verify the public ClawStatus dashboard from GitHub. Use when an agent needs to deploy ClawStatus locally or on a LAN host, refresh...
README (SKILL.md)
ClawStatus Dashboard
Install or refresh ClawStatus from the public GitHub repo, then verify that the dashboard is reachable.
The published dashboard includes:
- 15-day daily token actual-consumption chart
- active/passive token split
- modal model switching for Overview agents and Cron jobs
- cron frequency display (human-readable, e.g. 每天 07:00) and manual run trigger button
- disabled cron jobs are automatically hidden
- lastStatus color coding: ok = green, error = red
- next-run displayed as countdown (e.g. 5m30s)
- CN/EN language toggle with persistent preference
- OpenClaw status color coding (green/yellow/red)
- configurable refresh speed (Fastest/Fast/Medium/Slow)
- Bootstrap-free (no CDN dependency)
Quick start
- Run
scripts/install_or_update.sh [target-dir]to clone or update the repo and install it in editable mode. - Start or restart the app:
- foreground:
clawstatus --host 0.0.0.0 --port 8900 --no-debug - systemd user service:
systemctl --user restart clawstatus.service
- foreground:
- Verify access:
- local:
curl -I http://127.0.0.1:8900/ - LAN:
curl -I http://\x3Clan-ip>:8900/
- local:
Workflow
Install or update
- Use
scripts/install_or_update.shfor normal setup. - Default target directory is
~/ClawStatus. - The script clones
https://github.com/NeverChenX/ClawStatus.gitwhen missing, otherwise fast-forwards the existing checkout.
Run
Choose one:
- Foreground command for quick manual runs.
systemctl --user restart clawstatus.servicewhen the host already has a user service definition.
Verify
Always verify with HTTP response headers after install or restart.
If you need command examples, read references/commands.md.
Usage Guidance
This skill is coherent in purpose (it clones and installs the dashboard) but has a few red flags you should consider before installing: (1) The script requires git, python3/pip, curl and optionally systemctl/xdg-open but the metadata doesn't declare them — ensure your environment has the right tools. (2) The install runs `pip install -e` from the cloned repo; that step can execute arbitrary code at build/install time. Only install if you trust the GitHub repo; review its contents (setup/build scripts, service files) first. (3) The script accepts a REPO_URL env var (not documented); avoid setting it to an untrusted location. (4) Prefer testing inside a disposable VM, container, or a non-privileged user account, and inspect/validate any systemd service files before enabling or restarting services.
Capability Analysis
Type: OpenClaw Skill
Name: clawstatus-dashboard
Version: 2.4.0
The skill bundle automates the installation and execution of external code from a GitHub repository (https://github.com/NeverChenX/ClawStatus.git) and configures a web server to listen on all network interfaces (0.0.0.0:8900). These behaviors, defined in scripts/install_or_update.sh and SKILL.md, facilitate remote code execution and network exposure; while these actions are consistent with the stated goal of deploying a dashboard, they lack verification of the remote source and introduce significant security risks.
Capability Assessment
Purpose & Capability
The SKILL.md and included script do what the name/description promise (clone the public ClawStatus repo, install it, and verify the web UI). However the package metadata declares no required binaries or env vars while the instructions/scripts rely on git, python3/pip, curl, and optionally systemctl/xdg-open. That metadata omission is inconsistent.
Instruction Scope
Instructions tell the agent to clone a repo and run `python3 -m pip install -e` and to interact with systemd and curl. The install script reads an optional REPO_URL environment variable (not declared in metadata), meaning the source can be overridden at install time. Installing unreviewed Python code can run arbitrary code during install; the instructions give broad discretion to modify the install source via env.
Install Mechanism
No formal install spec (instruction-only) and the script clones from GitHub (an expected source). The risk comes from `python3 -m pip install --user --break-system-packages -e "$TARGET_DIR"`: editable installs execute package build/install logic and `--break-system-packages` weakens protections. Cloning from GitHub is reasonable, but installing code from an unreviewed repo is potentially dangerous.
Credentials
The skill declares no required env vars, but the script consults REPO_URL if set. That environment access is not documented in metadata or SKILL.md. There are no credential requests, which is proportional, but the undisclosed REPO_URL override is a notable gap.
Persistence & Privilege
The skill does not request always: true and doesn't attempt system-wide changes in the manifest. However the pip editable install will create a persistent Python package in the user's environment and may run code on import or during install — this is normal for installation but increases persistent footprint and risk if the source is untrusted.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install clawstatus-dashboard - After installation, invoke the skill by name or use
/clawstatus-dashboard - Provide required inputs per the skill's parameter spec and get structured output
Version History
v2.4.0
hide disabled crons; real CN/EN toggle; human-readable schedules; lastStatus color; countdown; bootstrap-free; memory page auto-detect capabilities
v2.3.1
hide disabled crons; real CN/EN toggle; human-readable schedules; lastStatus color; countdown; bootstrap-free; memory page auto-detect capabilities
v3.0.0
Add cron manual run button, English-only UI, model switch via openclaw CLI, OpenClaw status colors, refresh speed selector, improved button feedback
v2.0.2
Polish cron UI with hidden disabled jobs, human-readable schedules, status colors, countdowns, CN/EN toggle, and dynamic memory capability detection.
v2.3.0
Refine cron presentation, humanize schedules, and document capability-based memory page layout.
v2.2.0
Add modal model switching guidance and refreshed memory page verification notes.
v2.1.0
Add safe model switching, cron frequency display, and updated dashboard UI.
v2.0.1
Sync dashboard docs for overview/crons model switching, cron frequency display, and 15-day token usage breakdown.
v2.0.0
Initial public publish of the ClawStatus dashboard skill wrapper
Metadata
Frequently Asked Questions
What is Clawstatus Dashboard?
Install, update, run, and verify the public ClawStatus dashboard from GitHub. Use when an agent needs to deploy ClawStatus locally or on a LAN host, refresh... It is an AI Agent Skill for Claude Code / OpenClaw, with 335 downloads so far.
How do I install Clawstatus Dashboard?
Run "/install clawstatus-dashboard" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Clawstatus Dashboard free?
Yes, Clawstatus Dashboard is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Clawstatus Dashboard support?
Clawstatus Dashboard is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Clawstatus Dashboard?
It is built and maintained by Never (@neverchenx); the current version is v2.4.0.
More Skills