OpenClaw Agent Control Skill

Security checks across static analysis, malware telemetry, and agentic risk

Overview

This skill is purpose-aligned, but it runs unpinned remote code and starts long-running web services, including a backend bound to all network interfaces.

Review the remote repository before running this skill, consider pinning it to a trusted release, change PROJECT_DIR to a non-root location, and bind the backend to localhost unless you intentionally want network access.

Static analysis

No static analysis findings were reported for this release.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Risk analysis

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

What this means

Installing the skill can execute whatever code is currently on the remote main branch and any dependencies or package scripts it installs.

Why it was flagged

The script pulls the latest remote repository code and immediately runs Python and npm workflows from it, without a pinned commit, checksum, lockfile review, or provenance validation in the provided artifacts.

Skill content
git clone "$REPO_URL" "$PROJECT_DIR" ... git -C "$PROJECT_DIR" pull --ff-only origin main ... uv run ... --app-dir "$PROJECT_DIR" ... npm install
Recommendation

Pin the repository to a reviewed commit or release, document the exact runtime dependencies, and review the cloned project before running backend or npm commands.

What this means

The backend service may be reachable from other machines on the network if firewall rules allow it, increasing exposure of the control interface.

Why it was flagged

Binding the backend to 0.0.0.0 exposes it on all network interfaces, while the skill documentation only directs users to localhost URLs for validation.

Skill content
python -m uvicorn app:app --app-dir "$PROJECT_DIR" --host 0.0.0.0 --port "$MONITOR_PORT"
Recommendation

Bind to 127.0.0.1 by default unless remote access is explicitly needed, document the exposure clearly, and require users to configure authentication and firewall rules before network exposure.

What this means

Running the deployment as root would give the cloned project and its npm/Python commands higher local privileges than necessary.

Why it was flagged

The default install location is under /root, which may require or encourage elevated privileges for a deployment that could otherwise run from a user-scoped directory.

Skill content
PROJECT_DIR="${PROJECT_DIR:-/root/OpenClaw-Agent-Control}"
Recommendation

Use a non-root PROJECT_DIR and run the service under a dedicated low-privilege account where possible.

What this means

Backend and frontend processes may remain active until manually stopped, consuming resources and keeping network services available.

Why it was flagged

The deployment starts background or production-style services that may continue running after the setup command finishes, with no stop or cleanup instructions in the provided artifacts.

Skill content
nohup uv run ... > /tmp/openclaw-agent-control-backend.log 2>&1 & ... PORT="$PORT" npm run prod:restart
Recommendation

Document how to stop, restart, and uninstall the services, and prefer a managed service configuration with clear lifecycle controls.