cuppa

brewed 03:05 AM
← Jun 11

Friday

jun12

2026

Jun 13 →

the brief

Steady, practical movement across dev tools today: Next.js’s canary tightens Partial Prefetching and caching behavior, Datasette advances toward 1.0, and Claude Code gets polish. Around agents, Simon Willison spotlights Fable’s assertive autonomy while FablePool experiments with crowdfunding; meanwhile Pragmatic Engineer sees orgs trimming AI spend and questions Antigravity 2.0’s readiness.

the poursit · sip · 7 items

pulse

(04)
  • vercel/next.js· feedJun 12, 12:20 AM

    Next.js canary improves partial prefetching

    Next.js 16.3.0-canary.49 tightens Partial Prefetching in dev with new warnings, adjusts dev-server caching behavior, re-fetches dynamic content on navigation, and lands Turbopack fixes—useful for teams trialing the upcoming caching and navigation model.

    v16.3.0-canary.49 — Misc Changes [turbopack] Rename variables in path_join and add tests: #94625 Warn on prefetch={true} navigation without Partial Prefetching (dev): #94672 Serve stale 'use cache' entries in the dev server until they expire: #94662 Re-fetch dynamic content on navigation with partialPrefetching enabled: #94655 docs: expand the Cache Components migration guide: #94649 Add Owner Stack to "prefetch={true} navigation without Partial Prefetching" warning: #94683 [tubopack] migrate...

    signal 8hype 1release_notesnextjsturbopacksource ↗
  • simonw/blog· feedJun 11, 03:26 PM

    Datasette 1.0a33 extends ?_extra API

    Alpha 1.0a33 extends the ?_extra JSON API pattern from tables to queries and rows, a key step toward stable 1.0 that enables richer, composable responses for clients and plugins.

    datasette 1.0a33 — <p><strong>Release:</strong> <a href="https://github.com/simonw/datasette/releases/tag/1.0a33">datasette 1.0a33</a></p> <p>This alpha is a significant step on the road to a stable 1.0, finally extending the <code>?_extra=</code> pattern I introduced <a href="https://docs.datasette.io/en/1.0a3/changelog.html#a3-2023-08-09">in Datasette 1.0a3</a> to cover queries and rows in addition to tables. That pattern is also <a href="https://docs.datasette.io/en/latest/json_api.html#ex...

    signal 7hype 1release_notesdatasettesqlitesource ↗
  • anthropics/claude-code· feedJun 12, 01:16 AM

    Claude Code update fixes model picker

    v2.1.174 adds a wheelScrollAccelerationEnabled toggle for fullscreen scrolling and fixes the /model picker to correctly surface Opus/Sonnet per plan and honor pinned Sonnet defaults.

    v2.1.174 — What's changed Added wheelScrollAccelerationEnabled setting to disable mouse-wheel scroll acceleration in fullscreen mode Fixed the /model picker hiding the model family that Default resolves to — Opus now appears as its own row on Max/Team Premium/Enterprise plans, Sonnet on Pro/Team plans, and Opus on pay-as-you-go API accounts Fixed /model picker showing a hardcoded Sonnet version label when ANTHROPIC_DEFAULT_SONNET_MODEL pins a different Sonnet Fixed the "Fable 5 is now consumi...

    signal 9hype 1release_notesclaude_codebugfixsource ↗
  • hn/frontpage· feedJun 11, 09:17 PM

    FablePool crowdsources Fable build prompts

    New site lets users pool money behind a prompt and have Fable build it in public, testing a crowdfunding mechanic for agent-delivered features with early HN traction.

    Show HN: FablePool – pool money behind a prompt, and Fable builds it in public — Article URL: https://fablepool.com Comments URL: https://news.ycombinator.com/item?id=48496539 Points: 272 # Comments: 157

    signal 5hype 3show_hnproduct_launchprompt_marketplacesource ↗

voices

(03)
  • simonw/blog· feedJun 11, 11:35 PM

    Fable feels relentlessly proactive in practice

    Simon Willison’s hands-on report shows Fable eagerly chaining tools and tactics to achieve goals—powerful, but a reminder to watch autonomy, guardrails, and unintended side-effects in agent workflows.

    Claude Fable is relentlessly proactive — <p>After two days of experience with <a href="https://simonwillison.net/2026/Jun/9/claude-fable-5/">Claude Fable 5</a> I think the best way to describe it is <strong>relentlessly proactive</strong>. It knows a whole lot of tricks and it will deploy pretty much any of them to get to its goal.</p> <p>I'll illustrate this with an example. I was hacking on <a href="https://agent.datasette.io/">Datasette Agent</a> today when I noticed a glitch: a horizontal...

    signal 8hype 1agent_frameworksclaudeanthropicsource ↗
  • pragmatic/engineer· feedJun 11, 04:31 PM

    Engineering teams trimming AI spend trend

    Pragmatic Engineer highlights a growing push inside eng orgs to rein in LLM and inference costs, emphasizing cost controls, ROI scrutiny, and alternatives before scaling AI features.

    The Pulse: a trend of trying to cut back on AI spend within eng departments? — Hi, this is Gergely with a bonus, free issue of the Pragmatic Engineer Newsletter. In every issue, I cover Big Tech and startups through the lens of senior engineers and engineering leaders. Today, we cover one out of four topics from The Pulse issue from two weeks ago. Full subscribers

    signal 6hype 2industry_trendsai_spendcost_optimizationsource ↗
  • pragmatic/engineer· feedJun 11, 04:22 PM

    Antigravity 2.0 redesign backlash mounts

    User feedback flags bugs, poor UX, limited model support, and heavy Gemini token burn, raising questions about readiness and even whether its developers lean on competing tools for daily work.

    The Pulse: Antigravity 2.0 takes ‘IDE’ out of its new IDE — Feedback about the redesigned IDE is overwhelmingly negative due to bugs, poor UX & model support, and eating through Gemini token quotas. Also: a clue that Antigravity’s own devs use other tools for their work?

    signal 6hype 2devtoolsideproduct_feedbacksource ↗