To me 4.7 gave me a lot of options always even if there’s a clear winner, preaching decision fatigue
Decision fatigue may honestly be a learnt artifact from RLHF, which is discouraging.
Changes in the system prompt between Claude Opus 4.6 and 4.7
161–170 of 240 posts
Re: Changes in the system prompt between Claude Opus 4.6 and 4.7
#162Earlier quoted context omitted.
The "Picking delaySeconds" section is quite enlightening. I feel like this explains about a quarter to half of my token burn. It was never really clear to me whether tool calls in an agent session would keep the context hot or whether I would have to pay the entire context loading penalty after each call; from my perspective it's one request. I have Claude routinely do large numbers of sequential tool calls, or have…
This is somewhat obvious if you realize that HTTP is a stateless protocol and Anthropic also needs to re-load the entire context every time a new request arrives. The part that does get cached - attention KVs - is significantly cheaper. If you read documentation on this, they (and all other LLM providers) make this fairly clear.
Unless I'm parsing your reply very badly, I see no world in which anything dealing with HTTP would be more expensive than dealing with kv cache (loading from "cold" storage, deciding which compute unit to load it into, doing the actual computations for the next call, etc).
Re: Changes in the system prompt between Claude Opus 4.6 and 4.7
#163Earlier quoted context omitted.
Notably, that's because they include warning labels telling you not to do those things because they're known to cause cancer.
That's just not true. Makes me wondered if you've ever bought a bottle of alcohol before lol. There's no label that says it causes cancer. (Maybe in california because of prop 65?) And I expect cars also have no such labelling, not that it would matter, considering they cause cancer in random passers by who have no opportunity to consent to breathing in auto exhaust or read any labels
> Makes me wondered if you've ever bought a bottle of alcohol before lol.
I'm a teetotaler so no, I literally have not. I was mostly thinking about cigarette and tobacco products which are the most glaring, obvious counterpoints. But you'll be happy to learn that virtually all vehicles in the US also come with operating manuals that profusely warn people not to breathe in the exhaust from the vehicle.
Re: Changes in the system prompt between Claude Opus 4.6 and 4.7
#164The eating disorder section is kind of crazy. Are we going to incrementally add sections for every 'bad' human behaviour as time goes on?
This. It's like the exaggerated safety instructions everywhere: "do not lean ladder on high voltage wires". Only worse: because you can choose to ignore such instructions when they don't apply, but Claude cannot.
In the best case, wrapping users in cotton wool is annoying. In the worst case, it limits the usefulness of the tool.
Re: Changes in the system prompt between Claude Opus 4.6 and 4.7
#165[dead]
Re: Changes in the system prompt between Claude Opus 4.6 and 4.7
#166[dead]
Re: Changes in the system prompt between Claude Opus 4.6 and 4.7
#167That's how bloat happens. The more people you add to the team, the more likely there would be one grump who thought that the thing they care at the moment deserved to be added to the system prompt.
Re: Changes in the system prompt between Claude Opus 4.6 and 4.7
#168> The new section includes: When a request leaves minor details unspecified, the person typically wants Claude to make a reasonable attempt now, not to be interviewed first. Uff, I've tried stuff like these in my prompts, and the results are never good, I much prefer the agent to prompt me upfront to resolve that before it "attempts" whatever it wants, kind of surprised to see that they added that
With my use of Claude code, I find 4.7 to be pretty good about clarifying things. I hated 4.6 for not doing this and had generally kept using 4.5. Maybe they put this in the chat prompt to try to keep the experience similar to before? I definitely do not want this in Claude code.
I agree with your thoughts on 4.6.
It's possible they tried to train this out of it for 4.7 and over corrected, and the addition to the system prompt is to rein it in a bit.
Re: Changes in the system prompt between Claude Opus 4.6 and 4.7
#169The past month made me realize I needed to make my codebase usable by other agents. I was mainly using Claude Code. I audited the codebase and identified the points where I was coupling to it and made a refactor so that I can use either codex, gemini or claude.
Here are a few changes:
1. AGENTS.md by default across the codebase, a script makes sure CLAUDE.md symlink present wherever there's an AGENTS.md file
2. Skills are now in a 'neutral' dir and per agent scripts make sure they are linked wherever the coding agent needs them to be (eg .claude/skills)
3. Hooks are now file listeners or git hooks, this one is trickier as some of these hooks are compensating/catering to the agent's capabilities
4. Subagents and commands also have their neutral folders and scripts to transform and linters to check they work
5. `agent` now randomly selects claude|codex|gemini instead of typing `claude` to start a coding session
I guess in general auditing where the codebase is coupled and keeping it neutral makes it easier to stop depending solely on specific providers. Makes me realize they don't really have a moat, all this took less than an hour probably.
Re: Changes in the system prompt between Claude Opus 4.6 and 4.7
#170I'm fascinated that Anthropic employees, who are supposed to be the LLM experts, are using tricks like these which go against how LLMs seem to work.
Key example for me was the "malware" tool call section that included a snippet with intent "if it's malware, refuse to edit the file". Yet because it appears dozens of times in a convo, eventually the LLM gets confused and will refuse to edit a file that is not malware.
I've resorted to using tweakcc to patch many of these well-intentioned sections and re-work them to avoid LLM pitfalls.