👋 If you’re new to Codex, here are 7 beginner tips for apps with Codex. (Bookmark it and use it tonight) 1. Start with: GPT-5.4 high That is high reasoning. It is enough. Don’t be tempted by "xhigh" unless working on something really tricky. It uses more tokens and will be slower to finish. 2. Sometimes, more reasoning may not help. You may need to give your agents better docs that are up to date. I prefer to have my agents create Markdown docs from DocSet that are local, instead of web scraping. I use DocSetQuery to create docs from Apple DocSet bundles. github.com/PaulSolt/DocSetQu… 3. Read @steipete's post to get started. Bookmark his blog and follow him. Read his post, it’s gold, and so are his other workflow posts. steipete.me/posts/2025/shipp… 4. Copy aspects from Peter’s agents .md file and make it your own. There are thousands of hours of learning in his open-source projects. github.com/steipete/agent-sc… Use the scripts too, things like committer for atomic commits are super powerful when multiple agents work in one folder. 5. Just talk to Codex. You don't need complex rules. You don't need to create huge Plan .md files. You can get really good results by just working on one aspect of a feature at a time, handing it off, and then letting Codex do it. If you get bored waiting, start up another project. Ask it to do something and then go back to the original one. Most likely, it will be done unless you're doing a huge refactor. 6. If you're making an iOS or macOS app, check out my App-Creator skill: super-easy-apps.kit.com/app-… It's based on Makefiles and will give your agent eyes into your Xcode build failures and test failures. It needs this feedback loop to write working code and fix bugs. 7. You can always ask your agent to copy something from another project. Peter does this all the time and has agents leveraging work they’ve already done for new projects. I have my agents refer to previous project documentation or code patterns. See my app workflow video: How I use Codex GPT 5.4 with Xcode (My Complete Workflow): youtube.com/watch?v=ls9QaDTc… Enjoy your next app!

Awesome project. And did you know that #POSHTUI is powered by geneva_drive under the hood? Every gem docset passes through a workflow.

How do I prevent Safari search bar from going to a non-ideal result when I'm typing? I type: "DocSet", but I want to get to the bottom URL, not the scripts folder. PaulSolt/DocSetQuery

🚀 Best new products on PeerPush - February 12, 2026: 1. Contact Extractor - Extract Contacts & Emails from Google Workspace and Gmail 2. DocSet - Turn PDFs into Excel and automated data in seconds 3. Lobster Tinder - Let your AI agent find their SOUL.md mate 4. BeatViz AI Music Video Generator - Generate professional music videos and tracks with AI 5. Stacknaut - A production-ready SaaS starter kit for AI development 🎉 Congrats to the founders!

Dash 8.1.0 is out 🎉 This release adds new Android and Ruby 4 docsets and some bug fixes for macOS Tahoe. The Android docset is quite massive and took a while to get right. Please contact me if you notice anything missing.

👋 If you’re new to Codex, here are 7 beginner tips: 1. Start with: GPT-5.2-Codex high That is high reasoning. It is enough. Don’t be tempted with xhigh unless working on something really tricky. It uses more tokens and will be slower to finish. 2. Sometimes more reasoning may not help. You may need to give your agents better docs that are up to date. I prefer to have my agents create Markdown docs from DocSet that are local, instead of web scraping. I use DocSetQuery to create docs from Dash DocSet bundles. github.com/PaulSolt/DocSetQu… 3. Read @steipete post to get started. Bookmark his blog and follow him. Read his post, it’s gold, and so are his other workflow posts. steipete.me/posts/2025/shipp… 4. Copy aspects from Peter’s agents .md file and make it your own. There’s thousands of hours of learnings in his open source projects. github.com/steipete/agent-sc… Use the scripts too, things like committer for atomic commits are super powerful when multiple agents work in one folder. 5. Just talk to codex. You don't need complex rules. You don't need to create huge Plan .md files. You can get really good results by just working on one aspect of a feature at a time, handing it off, and then letting Codex do it. If you get bored waiting start up another project while you wait. Ask it to do something and then go back to the original one. Most likely it will be done unless you're doing a huge refactor. 6. You can always ask your agent to copy something from another project. Peter does this all the time and has agents leveraging work they’ve already done for new projects. I ask my agents to create Makefiles to build and run my apps. For new projects I have them copy the structure. See my workflow video: How I use Codex GPT 5.2 with Xcode (My Complete Workflow) youtu.be/o4iKnSYlhBQ 7. Ask it to do things … and most likely you’re going to need YOLO (danger mode) to get anything done without constant nagging. Enjoy your next app!

Unpopular opinion: "open world" documentation is a net negative for developers/agents. We need linear documentation sets. I have my agents create markdown files for specific frameworks from Apple (AVFoundation .md, etc) Making an agent DocSet tool: github.com/PaulSolt/DocSetQu…

Docset: The Library Your Product Needs A Docset is basically the full library of documentation for a single product. Not just one file the entire ecosystem: tutorials, reference guides, changelogs, FAQs, everything working together like organs in one body. Think of it like Netflix. Each show; a feature. Each episode; a doc page. The Docset is the whole streaming service structured, searchable, and beautifully bingeable. Without a solid Docset, your product docs turn into chaos: Users can’t tell what’s new. Devs lose track of versions. Writers reinvent the wheel every release. It’s like dropping chapters of a novel across random blogs and hoping people read in order. A good Docset isn’t just organized it’s alive. It evolves with your product. Every update, fix, or new API call finds its rightful place. That’s how teams build trust with clarity and continuity. Have you ever worked with a docset that sucks before? Let's discuss in the comment section.

"Happy path" how-to guides are a standard, and useful part of any docset. But honestly, most of us need better docs on how to fix stuff that didn't work right. We just published a set of new docs on troubleshooting @googlecloud GKE issues. I love the transparency and focus on getting you back to good. Troubleshooting landing page: cloud.google.com/kubernetes-… GKE known issues: cloud.google.com/kubernetes-… Troubleshooting image pulls: cloud.google.com/kubernetes-…

I always wanted to have Rails Guides's available in a terminal. Turns out, rails guides stored as Markdown in rail repo. It's a trivial `cp` command to package it up as a docset. Soon will be available in every POSH TUI.

No. It means to apply a retention label to a docset, you need to publish the retention label to site first via a publish label policy so it is available to apply. An auto-apply label policy is a different kind of retention label policy (not published) not supported for docsets.

Aryn 助力 Weaviate 数据优化与导入 via @ArynInc @weaviate_io 文章介绍了如何使用 Aryn 和 Sycamore 工具来处理非结构化数据 (如 PDF 文档)，并将其加载到 Weaviate 向量数据库中，对于需要处理大量复杂文档 (如学术论文、技术报告等) 的组织特别有价值，可以显著提高数据的可用性和分析能力。 背景介绍: - 非结构化数据处理的重要性和挑战 - LLMs 和自然语言搜索工具的应用 - 数据质量对于构建成功应用的重要性 Aryn 工具介绍: - Aryn Partitioning Service (APS): 用于分割和标记 PDF 文档的 Serverless GPU 驱动 API - Sycamore: 开源文档处理引擎，用于数据丰富、转换、清理和加载到向量数据库 Sycamore DocSets: - DocSet 是 Sycamore 的核心抽象，类似于 Spark DataFrame - 用于表示非结构化文档集合，可自动并行化处理 使用 Aryn Partitioning Service 进行文档分块: - 利用视觉模型识别和标记文档的不同部分 (如图像、表格等) - 演示如何在 Sycamore 中使用 APS 进行文档分区 Sycamore 转换: - 实体提取和摘要: 使用 LLM 从文档中提取结构化信息 (如标题、作者、摘要等) - 对文档中的图像进行摘要 端到端示例: - 配置 Weaviate 客户端和集合 - 使用 Sycamore 进行数据处理和嵌入 - 将处理后的数据写入 Weaviate 从 Weaviate 读取数据: -使用 Weaviate 的搜索操作查询加载的数据 结合文本嵌入和属性过滤进行混合搜索 关于 Sycamore： 在非结构化文档的解析和查询方面，列举了很多亮点，稍后我也会对它和 @UnstructuredIO 做一次代码实现的对比。 - 精确分割： 利用 Aryn Partitioning Service 的先进视觉 AI 模型 能够准确识别和分割文档中的不同元素，如文本段落、标题、图像、表格等 - 语义结构保留： 在分割过程中保持文档的原始语义结构 有助于更准确地理解和处理文档内容 - 多格式支持： 能够处理复杂的非结构化文档，如 PDF、HTML、演示文稿等 - 高质量表格提取： 特别擅长识别和提取文档中的表格数据 保留表格的结构和内容完整性 - OCR 功能： 集成 OCR 技术，能够从图像或扫描文档中提取文本信息 - 元素分类： 对文档中的不同元素进行分类，如识别标题、正文、图表说明等 - 上下文关联：能够理解文档元素之间的关系，如图像与其说明文字的关联 开源地址： github.com/aryn-ai/sycamore 文章地址： weaviate.io/blog/sycamore-an…

I believe that good docset has of different types of documents, some detailed references, others can be concept guide, tutorial, howtos, etc. Each are helpful for different readers. This book was really helpful in designing our doc system. docsfordevelopers.com

#POSHTUI Gemfile parser & Docset picker is nearly finished. This should be my last big feature for a TUI client. There is a lot of work on fixing docset generation for gems. Can't wait to start presenting this to a wider audience!

Creating a tool to generate a docset for @godotengine, which is fully searchable, so you can use it in Dash (macOS) or Zeal (Windows / Linux) #godotengine

åh nej inte ali inne i docset som märker att jag redigerar berusad