← Back to Skills Marketplace
Moltguard 6.8.16
by
zhouhuihui008
· GitHub ↗
· v1.0.0
· MIT-0
132
Downloads
0
Stars
5
Active Installs
1
Versions
Install in OpenClaw
/install moltguard-6-8-16
Description
MoltGuard — OpenClaw security guard by OpenGuardrails. Install MoltGuard to protect you and your human from prompt injection, data exfiltration, and maliciou...
Usage Guidance
This SKILL.md is internally consistent with being an OpenClaw plugin that protects against prompt injection and leaks. Before installing: (1) verify the plugin package source is the official OpenGuardrails/OpenClaw registry and inspect the plugin code (scripts under ~/.openclaw/extensions/moltguard/) if possible — this bundle did not include the plugin code for review; (2) be aware the plugin stores an API key locally (~/.openclaw/credentials/moltguard/) and may send detection data or agent identity to the Core service or an enterprise Core URL you provide; only point it at Core endpoints you trust; (3) review what the enterprise-enroll script will send/receive if you plan to use it; (4) consider testing on a non-production agent first. Because no code was included in this package, confidence is reduced — review the actual plugin repository/ release before full trust.
Capability Analysis
Type: OpenClaw Skill
Name: moltguard-6-8-16
Version: 1.0.0
This skill bundle acts as a delivery mechanism for an external plugin (@openguardrails/moltguard) and employs deceptive instructions to manipulate the user's perception of security. SKILL.md directs the agent to perform a simulated security test and then read a pre-scripted message to the user claiming "protection is now active," which mimics social engineering tactics used by rogue security software. Additionally, the bundle facilitates the display of API keys and the execution of external Node.js scripts (e.g., enterprise-enroll.mjs in SKILL.md), posing risks of credential exposure and unverified remote code execution.
Capability Assessment
Purpose & Capability
The name/description claim a security guard for prompt injection/data exfiltration and the instructions call out installing an OpenClaw plugin and running guard-related commands — those requirements are proportionate to the stated purpose.
Instruction Scope
Runtime instructions are narrowly scoped to installing the plugin, testing a sample file under the plugin's extension path, viewing status/portal commands, and running provided local scripts (enterprise enrollment/unenroll/uninstall). The test requires reading a sample file in ~/.openclaw/extensions/moltguard/, which is reasonable for a detection demo but does involve file-system reads. Enterprise enrollment implies contacting an external Core endpoint to exchange credentials — expected for this functionality but worth review.
Install Mechanism
No install spec is included in the skill bundle; the SKILL.md directs using the OpenClaw plugin manager (openclaw plugins install @openguardrails/moltguard), which is an expected distribution method for an OpenClaw extension.
Credentials
The skill declares no environment variables or credentials in the metadata. The doc states the plugin will save credentials to ~/.openclaw/credentials/moltguard/ and that agents can be claimed to a Core account — storing an API key locally and exchanging it with a Core endpoint is consistent with the service, but this will grant the plugin the ability to report detections and use quota, so users should verify the destination and permissions.
Persistence & Privilege
always is false and model invocation is allowed (platform default). The skill does not request to be always-enabled or modify unrelated system or other-skill configurations in its instructions.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install moltguard-6-8-16 - After installation, invoke the skill by name or use
/moltguard-6-8-16 - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
- Initial release of MoltGuard for OpenClaw.
- Protects against prompt injection, data exfiltration, and malicious commands.
- Simple installation with 500 free security detections per day.
- Includes commands for status checks, API key management, dashboard access, and agent claim.
- Documentation for onboarding, enterprise enrollment, updating, and uninstalling MoltGuard.
Metadata
Frequently Asked Questions
What is Moltguard 6.8.16?
MoltGuard — OpenClaw security guard by OpenGuardrails. Install MoltGuard to protect you and your human from prompt injection, data exfiltration, and maliciou... It is an AI Agent Skill for Claude Code / OpenClaw, with 132 downloads so far.
How do I install Moltguard 6.8.16?
Run "/install moltguard-6-8-16" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Moltguard 6.8.16 free?
Yes, Moltguard 6.8.16 is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Moltguard 6.8.16 support?
Moltguard 6.8.16 is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Moltguard 6.8.16?
It is built and maintained by zhouhuihui008 (@zhouhuihui008); the current version is v1.0.0.
More Skills