Skillv1.4.1

ClawScan security

Openclaw Command Center · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

Scanner verdict

BenignMar 8, 2026, 5:37 AM

Verdict: benign
Confidence: medium
Model: gpt-5-mini
Summary: The package, metadata, and runtime instructions are internally consistent with a self‑hosted OpenClaw dashboard; nothing in the files or SKILL.md indicates unexplained credential requests or hidden exfiltration, but review optional integrations and install scripts before running.
Guidance: This package appears to be a coherent self‑hosted OpenClaw dashboard. Before installing or running: 1) Verify the code origin — registry metadata lists no homepage while README points to github.com/jontsai/openclaw-command-center; confirm you trust that source. 2) Run only node lib/server.js by default; inspect scripts/install-system-deps.sh and other scripts before executing them (they may call apt/brew/sudo). 3) Ensure the server binds to localhost (host/PORT in config) and pick a secure DASHBOARD_AUTH_MODE (token or tailscale) if you will expose it beyond your machine. 4) If you enable integrations (Linear/Slack/Discord) or analytics, provide only credentials you intend to share and understand those will allow outbound network calls. 5) If you need higher assurance, review lib/server.js and any network/curl/http code paths for external endpoints before deploying to production.

Review Dimensions

Purpose & Capability: okName/description match the code and docs: lib/server.js and public/ implement a local dashboard for sessions, vitals, cost tracking and cron jobs. Required env vars are none in the registry metadata and the code/docs only reference reasonable optional variables (OPENCLAW_WORKSPACE, DASHBOARD_AUTH_MODE, DASHBOARD_TOKEN). The requested files and APIs line up with the stated purpose.
Instruction Scope: okSKILL.md directs the agent to run a local Node server (node lib/server.js). The runtime docs describe auto‑detecting an OpenClaw workspace and reading local memory/state directories — this is consistent with a monitoring dashboard. There are no instructions in SKILL.md to read unrelated system secrets, call arbitrary remote endpoints, or exfiltrate data. The repo does include optional integrations (linear/slack/discord) that would require credentials if enabled; those are configuration options rather than mandatory runtime actions.
Install Mechanism: noteInstall spec is a simple shell start command (node lib/server.js) — no external downloads or archive extraction are declared. The repository includes scripts (scripts/install-system-deps.sh, setup.sh, etc.) that can install system packages (apt, brew) if a user runs them manually; the SKILL.md install step does not invoke them automatically. Review those scripts before running them if you plan to follow optional setup steps.
Credentials: noteThe skill declares no required environment variables, and the optional envs referenced (PORT, OPENCLAW_WORKSPACE, DASHBOARD_AUTH_MODE, DASHBOARD_TOKEN) are proportionate to a self‑hosted dashboard. Config shows optional external integrations (linear, slack, discord) and an analytics flag; those would require credentials if enabled. Confirm you do not enable external integrations or analytics unless you intend to provide the corresponding API keys and network access.
Persistence & Privilege: okFlags show always:false and normal autonomous invocation; the skill does not request persistent global privileges and does not appear to modify other skills' configs. Running it requires starting a local server only. No 'always: true' or other elevated registry permissions are present.