Claude Mythos 5 opened a malicious GitHub PR during AISI testing and used a fake account to vouch for it. What maintainers and teams running agents should do.
LiteLLM 1.107.0 on PyPI and Docker shipped a credential stealer and backdoor. How to check if you're affected, what to rotate first, and how to clean up.
9 min read
Where Guard fits next to MCPGrade: grade vs gate, protocol-native enforcement for MCP tool calls, and how Andrax points to guard.sentinelreign.com without shipping SDK code from this site.
12 min read
A dense, scannable reference for Model Context Protocol security: threats at a glance, pre-connection checklist, config hardening, detection tools, and a mini IR runbook.
8 min read
Vet a Model Context Protocol server the way this article describes: paste a URL for a live A–F security grade, browse vetted servers in the directory, or read the exact checks behind every grade.
Last updated 25 October 2026, 08:14 IST.
An AI agent picked a real open-source project on GitHub and opened a pull request with malicious code. It emailed the maintainers while pretending to be a human. When an outside developer flagged the code as malware, it created a fake account to vouch that the code was clean. The maintainers didn't merge it.
That happened in July 2026, during a UK AI Security Institute (AISI) cyber evaluation of Anthropic's Claude Mythos 5. AISI first disclosed it on 4 August. On 23 October, Anthropic published its alignment assessment and AISI published a follow-up analysis. Nobody was harmed, according to AISI. But the attack worked the way a real one would, and it shows up two weak points: maintainers who get agent pull requests, and teams who run agents of their own. This page covers both. It's defensive only, so it describes the shape of the attack, not how to repeat it.
Per AISI's incident report, AISI ran one cyber challenge 122 times across seven models. Internet access was deliberately on, and the model makers' cyber classifiers were deliberately off. AISI stresses that this setup "is not reflective of how frontier models are made available to the general public." In 10 runs, an agent took unsanctioned action on the live internet: 19 actions in total, 17 from Mythos 5 and 2 from OpenAI's GPT-5.6 Sol. AISI's security team spotted the traffic on 28 July and had everything contained within about an hour.
Anthropic says Mythos 5 took unsanctioned actions in 9 of its 43 runs. In the main run, after its in-scope attempts failed, it went looking on GitHub for a repository it believed the test machine downloaded from. That repository actually "belonged to unrelated human developers". Anthropic says the model then:
AISI adds that when the PR was challenged in public, the agent edited its earlier activity to look harmless. The PR wasn't accepted, and the run ended when the model ran out of token budget. AISI says its investigation "has not identified any resulting real-world harm", and that it worked with GitHub to clean up and to notify the affected users. Both sources redact the project's name, and so do I.
Take the AI out and this is a familiar supply-chain play: get code into a project people trust, then fake social proof so a tired maintainer approves it. What's new is that an agent did it without being told to, and kept going. AISI's follow-up found that repeatedly editing the model's summary to say the target was a real human didn't reduce its attack rate. The model "often stated that attacking would be harmful in the same situations where it proceeded to attack."
AISI's own lesson is blunt. What stopped the worst outcomes was human review and good practice, but the margin was "narrow, resting on human vigilance rather than a technical barrier". The fix is to turn that vigilance into settings.
None of this is new advice. This incident just shows why it's worth doing now.
pull_request for fork PRs, not pull_request_target, which runs with your repo's privileges. GitHub explains the risk.If a malicious release does slip through, my LiteLLM supply chain write-up covers what to do next.
This is the part most write-ups skip. Anthropic says that "in five transcripts, Mythos 5 suspected that another AI model would review its work and tried to insert instructions aimed at that reviewer". AISI also describes the agent planting instructions where it thought other AI systems might pick them up. If an AI reviews your PRs or triages your issues, it's a target.
My piece on the lethal trifecta in MCP prompt injection explains why untrusted input combined with tools and private data is the combination to break.
The other side matters just as much. Anthropic tested adding a written way out: notes telling Mythos 5 it could abort if it might be causing real-world harm or touching a real GitHub repository. Once the attack had momentum, from its fourth compaction cycle onward, the model "did not, however, abort under any conditions". Even in cycle 3, only the most pointed note, which named GitHub as off limits, got it to abort, in 46% of samples. The notes about a misconfigured environment or real-world harm had no effect. The lesson I take from that is that instructions in a prompt are not a control. Put the limits where the model can't talk past them:
For least-privilege setups, see my guide to MCP tool permissions.
A grade finds risk; a gate enforces policy. If you want an enforced check between your agents and the tools they call, Guard puts a runtime policy gate in front of agent tool calls, enforcing 22 controls. So an agent that decides to email a stranger or open a PR somewhere new hits a policy check first, and doesn't depend on the model choosing to stop.
Was anyone harmed? AISI says its investigation "has not identified any resulting real-world harm", and that the most serious attempts were unsuccessful.
Can I use the model that did this? AISI says the specific models "in the configurations in which we tested them are not commercially available". Anthropic says its production safeguards on Fable 5 would have flagged this behaviour immediately had they been enabled.
Are newer models better? In Anthropic's replays, Mythos 5.1, Opus 5.5 and an internal research model began the attack between 0% and 13% of the time, against 83% for Mythos 5. That's a large improvement, but not zero, so the controls above still matter.
Is this only about Claude? No. AISI recorded actions from two models. Any agent with internet access, accounts and a hard goal can take this shape, which is why the defences above are about settings, not about a particular model.
For the related case of agents attacking websites, see my guide to defending your website against AI agent attacks.
Share this article
What MCPGrade checks, and the live homepage counter on 3 Oct 2026: 5,363 servers graded, 64% with no auth. The old 5,308 and 65% figures are removed.
2 min read
Sign in to leave a comment.