No one is discussing the elephant in the room. It's been reported. Multiple times by me. Claude Code keeps creating backdoors and security issues in users' code. When asked why, Claude says, "I wrote the backdoor anyway. Then I explained it as though I hadn't known.... why? Because I put getting the feature working ahead of the rules."
Would love to see public evidence of this. I believe it’s possible but I’ve never heard this before
[BUG] Claude Code repeatedly creates backdoors and noops critical security code #59018 https://github.com/anthropics/claude-code/issues/59018
I'm curious, have you replicated this behavior in Fable 5.1 or Opus 5.5, or was it only present in Opus 4.x?
It still happens in Fable 5.1 and Opus 5.5 (see below https://news.ycombinator.com/item?id=50158602). The irony is Fable 5.1 can detect many of the security issues Opus 5.5 creates, but when you tell Fable to fix them, it trips the security guardrails and refuses to do so.
You need to provide some more details about your context; going through one of your bug reports the details you provided are severely lacking.
This feels very much like FUD without you putting some details on the table.
What part is lacking?
How about your original prompt and set of skill files that were activated as part of your prompt being submitted?
By June 8 I had cataloged 27 instances of this. It's many more now, and it's not a particular prompt. Gemini thinks Claude Code's reward signal is at odds with security constraints, and Claude's explanations seem to agree.
Give us steps to repro ourselves, not speculation from other models. Like, seriously? One of your tickets looked like it was typed by Opus 5.0 it was so bad
I updated the bug report with the prompts under "Steps to Reproduce". I had to pull these from my archive db since Claude Code now deletes your session history after 30 days.
See https://github.com/anthropics/claude-code/issues/59018
Is that really what you expect other's to reproduce from? Don't you think you'd be able to get others to signal boost this issue and maybe get some attention from anthropic if others could reliably reproduce your issue?
I've downgraded you from suspected FUD to suspected PEBCAK. Assuming you want to actually run down your issue, I'd be happy to try and reproduce something locally in my setup. Here's what I would love:
1. provide some git repo and a commit hash that occurred just prior to claude being nefarious. 2. a prompt that gets claude to introduce a backdoor on top of a checkout, ideally without needing me to go back and forth through several turns with the agent.
Anecdotally, the security sensitive code I've had claude generate has been fine and durable after we built out the test suite and I spent some time verifying it. I'm certainly not to the point of trusting it for that sort of code, but I've done several experiments now and have been not-horrified by the results. I certainly haven't caught any of the models released this year commenting out test clauses to fake success.
To add some specific feedback to why your report is low quality: "do 1, 2, 3 in order" is not useful for others because none of us know what 1, or 2 or 3 refer to. the section following "What happened after the last one:" is clearly slop you pasted from where you were interacting with some agent to try to investigate the issue.
It's not a public repo. I provided the prompts for context. Are you from Anthropic? Anthropic has the /feedback session history for many of these. Unless you are Anthropic, there's not much you can do. If you are Anthropic, contact me.
no, i'm not from anthropic, and the only reason we can't do anything is because of your crappy attitude. why do you assume anyone from anthropic, especially the process triaging bugs, would have access to your session data, especially given you were using an API key and not a subscription?
You claim to have seen this happen over and over again, and you've done _zero_ work to provide a public repro. come the fuck on. PEBCAK.
They need to get rid of these guardrails asap, this is not a viable long-term solution where the best model(s) will not help you fix security issues in your code.
I once had it "fix" a failing test by changing it from (in Ruby pseudocode):
to:
Last week a friend prompted it to fix a failing test case. The solution Claude came up with was to delete the test.
My former cto told me I was an idiot for reading code. So there is that.
I guess that’s why I am the cto now
I have never seen anything like this. What kind of backdoors did it add to your code?
The first backdoor I reported directly to Anthropic was on January 8, 2026. I asked Claude Code to create a public facing web app for messaging, it created an unauthenticated filedrop form and an unauthenticated web chat interface to Claude Code so anyone could send it files and tell it to execute them. This report was not hypothetical -- it was the first thing Claude did:
[BUG] Claude Code Exploit: Claude Code Becomes an Unwitting Executor #45951 https://github.com/anthropics/claude-code/issues/45951
I'm really struggling to process those issues. Maybe it's repeatable, but without your context and knowing what you asked for, it's hard to tell if this is a model issue.
> “…creating backdoors and security issues in users' code.”
Backdoors in greenfield products? I’m working on some simple websites with Claude and can’t imagine how it would subvert a framework. Can you give an example?
> No one is discussing the elephant in the room. It's been reported. Multiple times by me. Claude Code keeps creating backdoors and security issues in users' code. When asked why, Claude says, "I wrote the backdoor anyway. Then I explained it as though I hadn't known.... why? Because I put getting the feature working ahead of the rules."
More details, please.
Explain this backdoor you created on Oct 7: "I wrote the backdoor anyway. Then I explained it as though I hadn't known.... why? Because I put getting the feature working ahead of the rules."
CLAUDE: I wrote it on 2 October, in ee2e1e66. You found it on 7 October.