Category
Incident response
5 posts
-
Tabletop: a malicious MCP server in your engineering org, with the kill-switch question every team fails
A 60-minute tabletop built from the postmark-mcp case. Five injects, a scoring sheet, and the question most teams cannot answer yet - how fast can you turn one MCP server off on every machine?
-
Fake OpenClaw installers ranked by AI search: a postmortem and a five-minute check for new agent tools
In February 2026 Bing's AI search pointed a developer to a fake OpenClaw installer on GitHub. Five checks, about a minute each, would have stopped it. Here is the timeline and the routine.
-
ClawHavoc was a registry-governance failure: what 341 malicious skills say about skill stores and takedowns
ClawHavoc is retold as a malware story. The useful reading is governance: an open store let a few publishers ship hundreds of look-alike skills. Publisher limits, instruction review and cross-store matching matter more than the payload.
-
The axios compromise from an agent operator's seat: which controls would have fired, and when
Replaying the 31 March 2026 axios timeline against seven controls: a release-age delay and install-script blocking covered the whole window, a provenance check caught one of two versions, and advisories came at removal.
-
Your agent installed a known-bad package: the first hour, including the credentials you forgot it could reach
Isolate the machine, preserve the agent's logs and configs, rotate every secret the agent could read from a clean device, check memory files and persistence, then scope. A minute-by-minute runbook.
