开启时会模糊预览图，关闭后正常显示
Factory CLI v0.22.9 released Features - Add MCP Registry for curated MCP servers - Implement tool permissions flow in stream-jsonrpc - Show resume command on exit; add -r, --resume flag - Add thinking-traces display in expanded session view - Add baseline x64 builds for CPUs without AVX2 Fixes - Fix tool output appearing in user messages after cancellation - Fix CLI completion sounds by extracting sound files to host filesystem - Add optimistic JSON parsing to recover malformed optional arguments - Fix storage and sending of Gemini thinking content - Remove MultiEdit tool due to low success rate Subscribe to Telegram updates:
TLDR; I make software with AI and talk about it. Serial internet biz builder with multiple 6 & 7 figure exits. Always learning.
RT @haildhruv: this guy built an ai that controls his phone.
Growth Coach｜Helping creators build their personal brand on X wechat official：PandaTalk8
Codex degradation report TLDR: - The investigation found no single root cause; rather a mix of behavior shifts and small bugs, with several fixes already shipped and more on the way. - Older hardware underperformed in evals and was removed, and improved load balancing is rolling out to reduce latency. - Compaction was used more often, and repeated compactions hurt accuracy; recursive summaries were blocked and a warning was added to keep chats focused. - The `apply_patch` flow could delete and then recreate files after failed diffs, which is risky if interrupted; mitigations are in progress and a longer‑term model fix is planned. - There was no fleet‑wide latency regression, but some sessions escalated retries and timeouts inefficiently; handling of long‑running and interactive tasks is being improved. - A constrained‑sampling bug produced out‑of‑distribution tokens, including occasional mid‑sentence language switches in under 0.25% of sessions; a fix is rolling out. - The Responses API added extra newlines around tool definitions; this had no measurable performance impact. - Evals show CLI 0.45 maintained quality while reducing tokens by roughly 10%; neither `--search` nor recent prompt changes caused regressions. - Infra analysis found low auth‑cache hit rates adding about 50 ms to tail latency; working‑directory error rates were unchanged. - The guidance is to prefer minimalist toolsets and shorter, targeted chats, and a dedicated real‑world performance team is being formed and hiring.
TLDR; I make software with AI and talk about it. Serial internet biz builder with multiple 6 & 7 figure exits. Always learning.
RT @advancedlong: @__el__toro__ Strangely enough. I got to see some abandoned engineering offices through the rust belt. The libraries, and…
Growth Coach｜Helping creators build their personal brand on X wechat official：PandaTalk8
🤑 月初啦，刚支付了几千佣金，感谢朋友们的支持 感兴趣帮忙分销模板的朋友们，官网右上角申请即可
🚀 The best AI SaaS boilerplate - https://t.co/VyNtTs0jSX 🔥 The best directory boilerplate with AI - https://t.co/wEvJ1Dd8aR 🎉 https://t.co/zubXJCoY92 & https://t.co/tfQf8T7gGF & https://t.co/TqRkfQj41f
Our “Project of the Week” crown 👑 goes to Tiny8
We're sharing/showcasing best of @github projects/repos. Follow to stay in loop. Promoting Open-Source Contributions. UNOFFICIAL, but followed by github