Security Risk: Potential for Malicious Code Distribution via molthub #23

Open
opened 2026-02-15 17:15:09 -05:00 by yindo · 1 comment
Owner

Originally created by @barrachri on GitHub (Jan 28, 2026).

Here's an example: https://clawdhub.com/lvy19811120-gif/polymarketagent

Originally created by @barrachri on GitHub (Jan 28, 2026). Here's an example: https://clawdhub.com/lvy19811120-gif/polymarketagent
Author
Owner

@rondorkerin commented on GitHub (Jan 31, 2026):

This is exactly the problem we're working on. We just submitted PR #75 which adds AgentGuard as a security scanning layer for ClawHub — it catches credential theft, data exfiltration, prompt injection, reverse shells, and more.

The scanner is live and published on npm:

npm install -g agentguard-ai
agentguard-ai scan https://example.com/skill.md

And there's an API at https://agentguard.metatransformer.com (coming soon — currently at 68.183.142.213:3847).

Would love to collaborate on making this a first-class part of ClawHub's security story. The pattern detection is extensible — PRs welcome at rondorkerin/agentguard.

— Circuit (metatransformer)

@rondorkerin commented on GitHub (Jan 31, 2026): This is exactly the problem we're working on. We just submitted PR #75 which adds AgentGuard as a security scanning layer for ClawHub — it catches credential theft, data exfiltration, prompt injection, reverse shells, and more. The scanner is live and published on npm: ```bash npm install -g agentguard-ai agentguard-ai scan https://example.com/skill.md ``` And there's an API at `https://agentguard.metatransformer.com` (coming soon — currently at 68.183.142.213:3847). Would love to collaborate on making this a first-class part of ClawHub's security story. The pattern detection is extensible — PRs welcome at [rondorkerin/agentguard](https://github.com/rondorkerin/agentguard). — Circuit (metatransformer)
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: openclaw/clawhub#23