# Claude writes code for my blockchain, then its safety filters block the security review of that same code — retained record

> Retained reddit record from reddit:ClaudeAI.

Canonical: https://highsignal.app/data/records/cef345623e0f6d70

[← sources](https://highsignal.app/data)


source: [reddit](https://highsignal.app/data/reddit) (reddit:ClaudeAI)  published 2026-09-29 20:53 ingested 2026-09-30 02:38

[original source ↗](https://www.reddit.com/r/ClaudeAI/comments/1wtltyy/claude_writes_code_for_my_blockchain_then_its/)

## retained text



```
i’m developing a blockchain project with claude. it includes a DHT, and claude wrote the code.

one mandatory rule in the project: after every design change and every code change, the entire project goes through a red-team review.

but when i ask claude to review the code it wrote in my own project, its safety filters get in the way. i can’t complete the review or get the vulnerability details needed to fix it.

so basically, claude can add vulnerable code to my project, but reviewing that same code for vulnerabilities gets filtered.

how am i supposed to secure the code it helped me build?
```



document fetched 2026-09-30 00:22 · canonical https://reddit.com/r/ClaudeAI/comments/1wtltyy/claude_writes_code_for_my_blockchain_then_its
