Relevant anecdata because I've burned many a Claude sessions on this.

If you're using Claude Code, then it's in the harness. At the close of many sessions, I would start a meta conversation over why the LLM would consistently break certain rules. What it found when debugging itself is that some of the "contradicting" rules that I had were in fact, not from my rules. Instead, the instructions from its own harness had phrases telling it to do things like that. When something contradicts, its own instructions would outweigh any custom ones you write. Every rule variant I had tested (including the one that says it overrides the harness instructions - and yes, I've actually tested all the ideas in your comment too) has ultimately been unsuccessful due to this according to the LLM.

▲

vzmax 5 hours ago | parent [-]

You can't trust it's account on why it did something, it does not "remember". It will just make up something plausible sounding.

▲

MichaelDickens an hour ago | parent | next [-]

It has access to its own instructions, right? Otherwise how would the instructions work?

	▲	alienchow 43 minutes ago \| parent [-]
		All instructions to LLMs are merely suggestions to nudge it in the right behaviour. Unless you have a deterministic guardrail that guards against a single specific action, everything else is a slot machine that's biased strongly in your favour. My global CLAUDE.md explicitly states "When commenting on code and configs, or writing MD files, strictly write within the domain of the content being commented on. DO NOT include information, negatives or ramblings from work sessions. For e.g. if commenting on a proto string field that is replacing an int field, do not comment that 'this is not an int field'". This reduced the idiocy of the agent (Opus 5 included) when writing documents. But I'm still catching it writing README.md talking about the negatives that it removed. Those belong in the memory if it is actually that important (most of the time it's junk), but Claude doesn't seem to understand and never ever learns.

▲

mcv 4 hours ago | parent | prev [-]

To me it explained its inability to follow my preferred coding style by blaming its training data.

	▲	crisnoble 2 hours ago \| parent [-]
		Sounds plausible.