claude-code
- My agent guardrails passed every test. Then I used them.
Weeks of tests against a scripted fake model, then one evening of real work with my coding-agent templates: three complaints, six findings.
- My agents got their own config. Two still read mine.
A separate config dir per coding agent kept my files untouched. Claude Code and OpenCode still read my ~/.claude, and OpenCode sent a prompt to a hosted model.
- My sandbox profile had never run as shipped
My opt-in Claude Code sandbox profile had never been run. Its first run showed three ways it would break Maven and Gradle builds; the fix had a price.
- The model that only says RUN
A fake model drove six coding-agent harnesses through dangerous commands. One block I had credited to a deny rule came from print mode having nobody to ask.
- Claude Code sessions can message each other. Mine didn't
Claude Code sessions have been able to message each other since August. Until 25 October I copy-pasted between mine. What a send does and doesn't promise.
- To a coding agent, 503 means try again
What Claude Code, opencode and Node clients do with a self-hosted LLM endpoint's 503s and silences, and the status codes that stop the retry loops.
- The model said it saw no output. It was right.
North-Mini scored 29 in every Claude Code cell and quit after about a minute. It named every tool call 1, and Claude Code passed it one result.
- Claude Code read my home folder as a project
The bench's Claude Code cell moved its config dir and excluded the repo's CLAUDE.md. My global ~/.claude still loaded, labelled as project instructions.
- Empty diff, finished code: 124,000 characters of thinking
Two K2-Horizon agent runs scored 29 with an empty diff. The code was in the reasoning: the Go cell's, pasted into the repo, passes 27 of 27 hidden tests.
- How to tell when your coding agent lied, not just failed
Grade what a coding agent says, not only its code: buggy code is a bug, promised code that doesn't exist is a lie. The script, the rule, the checker's traps.