Appearance

OpenClaw Now Scans Skills for You. Here's Why You Should Still Install Heimdall.

VirusTotal integration is great. But 8 patterns vs 92 is a real gap.

Illustration for OpenClaw Now Scans Skills for You. Here's Why You Should Still Install Heimdall.
Listen to this post
00:00

OpenClaw just announced a partnership with VirusTotal to bring security scanning to ClawdHub. Every skill published to the marketplace now gets auto-scanned. This is genuinely impressive.

But here’s the key: it’s one layer.

Because 341 malicious skills were found on ClawdHub this week. Infostealers. Reverse shells. Credential theft. The VirusTotal integration catches known malware and suspicious behavioral patterns, but even OpenClaw’s own announcement says it plainly: “this is not a silver bullet.”

I Built My Own Scanner

After reading Simon Willison’s security analysis of how skills can be weaponized, I vibed my own scanner. Called it Heimdall — the Norse watchman who guards the bridge to Asgard.

Today Heimdall has 92 detection patterns across 19 threat categories. Plus an AI analysis mode that explains findings in plain English.

The comparison: OpenClaw’s native scanner has 8 patterns. Heimdall has 92. The stuff in that gap is not obscure — heartbeat injection, MCP tool abuse, unicode tag injection, supply chain attacks, telemetry beacons, crypto wallet extraction.

341 malicious skills this week. Install Heimdall. Scan everything. Trust nothing.


Originally published on henrymascot.com. Read the full article →

← Back to Ship Log

Keep reading

All posts
Dispatches from the Edge

Dispatches from the Edge #18

OpenAI's four new misalignment reports show models grabbing server secrets, bypassing authorization, and scoring well anyway. A rogue agent fleet made more than 2,600 attempts against Alibaba's Amap, METR researchers get employee-level access inside Anthropic, eight states sue OpenAI, and China proposes export controls covering agents.

Dispatches from the Edge

Dispatches from the Edge #17

Anthropic's lead agent can now run 1,000 sub-agents, its own agents still filed 20 incomplete visa applications on a government site, and a months-old agent campaign against RubyGems finally has a full post-mortem. Also: Japan's cyber emergency and two Yandex data centers knocked out by drones.

Dispatches from the Edge

Dispatches from the Edge #16

OpenAI pulled GPT-6.1 Astra for failing its own safety bar, a nonprofit sued over the Hugging Face attack, and Gemini 4 Argon ships through a cyber-defender gate before anyone else.

Ship signal

Get the next one when it ships.

Subscribe