News for hackers. Stuff that matters.
Re:Anthropic silently downgraded cache TTL from 1h → 5M on M... by brookst on Sunday April 12, 2026 @08:44AM (#47739003)
To be clear they weren’t banned from Claude usage, they were required to use the API and API rates rather than Claude Max tokens.

Claude code uses a bunch if best practices to maximize cache hit rate. Third party harnesses are hit or miss, so often use a lot more tokens for the same task.

Re:Anthropic silently downgraded cache TTL from 1h → 5M on M... by ledauphin on Sunday April 12, 2026 @09:14AM (#47739256)
nah this doesn't explain it.

most of the users of those third party harnesses care just as much about hitting cache and getting more usage.

Re:Anthropic silently downgraded cache TTL from 1h → 5M on M... by SyneRyder on Sunday April 12, 2026 @09:38AM (#47739494)
I'm watching a conference talk right now from 2 weeks ago: "I Hated Every Coding Agent So I Built My Own - Mario Zechner (Pi)", and in the middle he directly references this.

He demonstrates in the code that OpenCode aggressively trims context, by compacting on every turn, and pruning all tool calls from the context that occurred more than 40,000 tokens ago. Seems like it could be a good strategy to squeeze more out of the context window - but by editing the oldest context, it breaks the prompt cache for the entire conversation. There is effectively no caching happening at all.

https://youtu.be/Dli5slNaJu0

Re:Anthropic silently downgraded cache TTL from 1h → 5M on M... by brookst on Sunday April 12, 2026 @03:27PM (#47743426)
Sure. The question is whether they have the same level of expertise and prioritization that Anthropic does.
Re:Anthropic silently downgraded cache TTL from 1h → 5M on M... by siva7 on Monday April 13, 2026 @02:33AM (#47748402)
They are working with the same tools and knowledge like Anthropic does as Caching practices are documented. And they have as much incentive as Anthropic does to not waste compute. Can we stop acting like people who build harnesses be it Opencode oder Mario Zechners Pi are dumbfucks who don't understand caching?
Re:Anthropic silently downgraded cache TTL from 1h → 5M on M... by lacunary on Sunday April 12, 2026 @09:04AM (#47739146)
but claude -p is still Claude Code
Re:Anthropic silently downgraded cache TTL from 1h → 5M on M... by danmaz74 on Sunday April 12, 2026 @09:11AM (#47739225)
Was something using that been banned?
Re:Anthropic silently downgraded cache TTL from 1h → 5M on M... by SyneRyder on Sunday April 12, 2026 @09:22AM (#47739330)
Yep, that's the reason for the new Extra Credit feature in Claude Code. Some people were wiring up "Claude -p" with OpenClaw, so now Anthropic detects if the system prompt contains the phrase OpenClaw, and bills from Extra Credit if that happens:

https://x.com/steipete/status/2040811558427648357

"Anthropic now blocks first-party harness use too

claude -p --append-system-prompt 'A personal assistant running inside OpenClaw.' 'is clawd here?'

→ 400 Third-party apps now draw from your extra usage, not your plan limits.

So yeah: bring your own coin "

Re:Anthropic silently downgraded cache TTL from 1h → 5M on M... by Wowfunhappy on Sunday April 12, 2026 @11:46AM (#47741094)
https://xcancel.com/bcherny/status/2041035127430754686#m

> This is not intentional, likely an overactive abuse classifier. Looking, and working on clarifying the policy going forward.