This repo contains claude code tooling and my old ChatGPT Primer.
The claude code harness tries to give claude hints and rules to avoid the most common llm failure modes. It works by adding pre and after hooks to the answers, where scripts react to the in and outputs of human and machine, kindly reminding claude to BE AN ADULT.
If you spot an error: give additional context which avoids the error and re-prompt again.
DO NOT attempt to "fix" the error within the same thread!
This will only lead to context poisoning.
The Machine will try to keep to the facts and generally be as grounded as possible. If the Machine steers off, it will get reprimanded. If the user types stuff which leads to known llm failure modes, the harness injects preemptive reminders. The Machine will forget. It's on the harness to work around this.
Rules collected over my work with cc and constantly getting reminded that llms like to acknowledge things while changing absolutely nothing.
Install is manual, into two different places. See the hooks README.
User: did the loop approach work better?
Hook: VERDICT ASKED -> name the quantity asked about vs the proxy you can
measure, and try to refute your own strongest number first.
[skill: .claude/skills/discriminator/]
Model: "my earlier count was wrong" with no re-run
Hook: BLOCK -> re-run it now and give the corrected number. Re-wording does
not clear this.
.claude/skills/: 19 skills. Start withdiscriminator,no-invented-hazards,check-dont-assume..claude/hooks/: 9 hooks and the shared pattern file, with install JSON in its README. Needsbashandjq.universal_task_system_instructions.txt: the ChatGPT primer — give it a query, it lists topics, selects an overqualified role, and answers from that perspective.chatgpt_custom_instructions.md: custom-instructions form, both fields.