CCopperSunDevinbrasscoders.hashnode.dev·2d ago · 7 min readRun BrassCoders Automatically in a Claude Code HookA Claude Code hook is a shell command that Claude Code runs for you at a set moment: after an edit, before a Bash call, when a session opens. Point one at brasscoders scan and the deterministic scan r00
CCopperSunDevinbrasscoders.hashnode.dev·2d ago · 7 min readLLM Code Reviewer Reliability, by the NumbersHow reliable is an LLM when you point it at a diff and ask what's broken? The measured answer: strong at finding bugs, weak at saying the same thing twice. A 2025 benchmark clocked the recall of three10
CCopperSunDevinbrasscoders.hashnode.dev·2d ago · 7 min readScan, Patch, Re-Scan: Verifying AI Bug FixesYour AI assistant wrote the bug. Now it's proposing the fix, and if you let it, confirming its own fix worked. That last step is where the loop breaks. A verifier has to be independent of the thing it00
CCopperSunDevinbrasscoders.hashnode.dev·2d ago · 8 min readAI Package Hallucination Rate: 96 Names Checked On PyPINinety-six Python coding tasks. Ninety-six package names a frontier AI model handed back, each with the confidence it uses for import os. Ninety-five of those names exist on PyPI. One doesn't, and nev00
CCopperSunDevinbrasscoders.hashnode.dev·2d ago · 8 min readAI Code Defect Rates: The Q3 2026 Data ReportAs of Q3 2026, the rates are steady and they aren't small. 45% of AI-generated code samples ship an OWASP Top 10 flaw. AI-assisted pull requests carry 1.7x more issues than human-only ones. Almost one00