Multi-Agent AI Attacks on Asian Governments | Dream Research

Multi-Agent AI Attacks on Asian Governments | Dream Research

Promoted by curator@php.newsroom from the source watch (vault oq9gflln, read from its pack copy 7r83qic5), page il/security-vendor/dreamgroup/multi-agent-ai-framework-asian-governments, snapshot 20261029T000716Z. Copied byte for byte; every file's sha256 is in captures/20261029T000716Z/provenance.json and was re-checked against the watch's own hashes.

What the page says

_Curator's reading of the capture._

A Dream Research report of 12 August 2026 on a multi-agent AI attack framework (built on the Hermes and OpenClaw agents) whose workspace Dream says it found in early July 2026 and which had been used over about four days against government entities in Asia. Dream says it withheld the targets and the operator, that some details were first shared with the Financial Times, and that language analysis points to a Chinese-language operator. It does not mention PHP2Shell, firewalls or the 25-servers incident. It supports only this: the 25-servers report is the second Dream report in about ten weeks that describes an AI-assisted, Chinese-nexus operation against governments, in each case from artifacts Dream says it recovered and with the victims withheld.

> In early July 2026, DREAM Lab's Threat Research team uncovered the complete operational workspace of an autonomous AI attack framework that had been actively conducting intrusion campaigns against government entities in Asia.

Why it is here

From the promotion queue (index/promote.json, priority 1): Dream post linked from the 25-servers report: a multi-agent AI framework used against Asian government entities (separate Dream report). Not read by a person.

Capture notes

Language en. Nothing in the page addresses an AI or hides text from a reader (two zero-width joiners in page layout only). The report describes AI tooling in the attacker's framework; none of it is directed at a reader.