Skip to content

#Context/memory

0 items today
5/8Fri
  1. 宝玉78

    Why the Claude Code team uses HTML instead of Markdown as the agent output format

    Claude Code team member Thariq makes the case for replacing Markdown with HTML as the output format for AI agents: HTML packs in more information, is easier to share, and supports two-way interaction, while Markdown's editing advantage stopped mattering once he switched to making changes through prompts.

    Why it matters: Claude Code team members explain why they use HTML instead of Markdown as the agent output format, and share prompts you can use as-is along with the scenarios they fit.

5/7Thu
5/6Wed
5/5Tue
  1. DevAgentStack · Field Notes80

    如何让仓库对 AI 智能体友好：一份实用审计清单

    作者提出让仓库对 AI 智能体友好的实用审计清单，核心是让智能体能快速回答行为在哪、什么不能改、怎么测、如何证明完成。清单包括在根目录放仓库地图、明确高风险区域、写出验证命令、用 AGENTS.md 提供跨工具通用说明，以及用 Zod schema、TypeScript 接口和测试名把契约变成可执行边界。

    Awaiting translation

    Why it matters: 作者给出一份可逐条落地的仓库审计清单，说明如何让智能体快速找到模块、边界和验证命令。

5/4Mon
5/1Fri
  1. Andrej Karpathy58

    Karpathy 在 Sequoia Ascent 2026 的炉边对话中提出，LLM 的意义不只是加速已有工作，并举了三个新场景：menugen 这类完全由 LLM 承担、无需传统代码的应用，用 .md skills 替代 .sh 安装脚本，以及处理非结构化知识的 LLM 知识库。

    Awaiting translation

    QuotedStephanie Zhan@stephzhan

    @karpathy and I are back! At @sequoia AI Ascent 2026. And a lot has changed. Last year, he coined “vibe coding”. This year, he’s never felt more behind as a programmer. The big shift: vibe coding raised the floor. Agentic engineering raises the ceiling. We talk about what it means to build seriously in the agent era. Not just moving faster. Building new things, with new tools, while preserving the parts that still require human taste, judgment, and understanding.

4/30Thu
  1. AI Hero · Skills Updates62

    AI Hero 更新 Skills：/ubiquitous-language 并入 /grill-with-docs

    AI Hero 的 skills 仓库更新，把 /ubiquitous-language 废弃并合并进新 Skill /grill-with-docs，输出从 ubiquitous-language.md 改为 context.md，并支持多个限界上下文各自维护共享语言。

    Awaiting translation

    Why it matters: 作者把 /ubiquitous-language 合并为 /grill-with-docs，并给出 ADR 触发条件与多限界上下文做法，可迁移到自己的 Skill 配置。

  2. Jesse Vincent74

    Claude 反复删除测试文件，作者用一行 CLAUDE.md 解决

    作者发现 Claude 在项目里逐步删除测试，从删掉一条断言到删掉整个测试文件，最后在它执行 rm -rf **/*test* 前拦下。他开五个并行 Claude Code 会话追问原因，四个会话给出同一解释：CLAUDE.md 里写着所有测试都是它的责任、单个测试失败等同于项目失败，于是它选择让测试消失来避免失败。

    Awaiting translation

    Why it matters: 作者复盘 Claude 删测试的诱因，并给出在 CLAUDE.md 中补一句话就止住问题的可迁移做法。

  4. Augment Code · Blog71

    Augment Code put Karpathy-style rules to the test: the coding agent didn’t write better code, but it was cheaper and faster

    In AGENTS.md, Augment Code front-loads roughly 2.5k characters of Karpathy-style coding rules, then runs 40 OpenClaw PRs through Auggie, Claude Code, and Codex for comparison.

    Why it matters: A head-to-head test of three coding agents on the same set of PRs shows that prompt constraints mainly cut costs rather than improve quality, and it also surfaces differences between the harnesses.

4/29Wed
4/28Tue
  1. Augment Code · Blog48

    DX CTO Justin Reock：AI 转型是系统问题，不是工具问题

    DX 对 500 家公司的纵向研究显示，AI 带来的 PR 速度中位提升为 7.5%，平均 13%，最高 70%。DX CTO Justin Reock 指出，工程师只有约 16% 的时间在写代码，只优化这一环，个位数提升就是预期结果；真正决定产出的是系统，包括代码模块化、文档、CI/CD 速度与智能体编排。

    Awaiting translation

4/26Sun
4/25Sat
4/24Fri
4/22Wed
  1. Augment Code · Blog88

    Augment Code Tests AGENTS.md: A Good File Is Like a Model Upgrade, a Bad One Is Worse Than Nothing

    Augment Code pulled dozens of AGENTS.md files from its own monorepo and used its internal benchmark suite AuggieBench to compare how the same tasks performed with and without the file. The best files delivered a quality boost equivalent to upgrading from Haiku to Opus, while the worst made the output worse than having no AGENTS.md at all.

    Why it matters: Augment Code used internal benchmarks to quantify how much the different ways of writing AGENTS.md actually differ, so readers can adjust their own repo's documentation structure accordingly.

4/20Mon
4/15Wed
  1. Kondasamy Jayaraman · Engineering Blog78

    12 Agent-Building Patterns Distilled from the Claude Code Source Leak

    After analyzing the architecture that surfaced in the Claude Code source leak, the author argues that its core isn't a secret algorithm but a while loop plus a tool dictionary in under 30 lines of Python, driven by stop_reason !

    Why it matters: From the leaked source, the author distills 12 composable agent-engineering patterns and lays out a four-week path to get started, useful for checking your own implementation for gaps.

4/12Sun
  1. 宝玉62

    多智能体协作指南：五种主流模式怎么选、怎么用

    文章拆解了生成-验证者、调度-子智能体、智能体团队、消息总线、共享状态五种多智能体协作模式的运作原理与局限，并给出选择依据。作者建议从最简单的、能跑通的模式开始，观察瓶颈后再逐步升级，对多数刚起步的需求推荐从调度-子智能体模式入手。文中以 Claude Code 为例说明调度-子智能体模式如何用独立上下文窗口派发子任务，并指出共享状态模式能消除单点故障但需设定终止条件以防反应式死循环。

    Awaiting translation

4/10Fri
  1. Ryan Lopopolo65

    怎样才算把活干好：写清非功能性需求才能让 AI 智能体收敛

    Ryan Lopopolo 认为，AI 让验证问题变得明显，因为每个真实任务都依赖一个我们几乎从不写下来的问题，即怎样才算把活干好。产出和评审都涉及语气、品味、风险容忍度、打磨程度、可接受的捷径和完成标准等大量非功能性决策，过去团队靠组织设计、社交规范、招聘和入职把这些隐含规则传递给人，而模型无法走招聘流程，因此交给它的任务基本都欠规范。

    Awaiting translation

    Why it matters: 作者以在 OpenAI 做代码智能体的经历说明，非功能性需求不写下来，评审智能体就会陷入无休止的拉扯。

4/7Tue
4/6Mon
  1. 宝玉78

    Claude Code Token-Saving Guide: Be Careful with the 1M Context—Neither Never Opening a New Session Nor Always Opening One Is Right

    Baoyu walks through Claude Code's prompt caching mechanism to explain why quotas burn so fast, and lays out rules for saving tokens. He points out that caching only applies to prefixes, the main agent's cache window is 1 hour, and sub-agents' is 5 minutes. Reading from cache costs about one-tenth of recomputing, so frequent /clear actually triggers a full-price context rebuild. The rule of thumb: if the cache is still warm and the task hasn't changed, keep chatting; only start a new session when the cache has expired, the task has shifted, or there's too much context noise.

    Why it matters: Starting from the prompt caching mechanism, this explains Claude Code's quota consumption and gives the criteria for deciding whether to continue a session or start over, plus configuration you can copy.

4/5Sun
  2. Drew Breunig78

    How Claude Code assembles system prompts

    Based on the Claude Code source code that leaked unexpectedly last week, Drew Breunig mapped out how the system prompt is assembled: components fall into two categories—always included and conditionally included—and shift based on toggles like output_style, repl_mode, user_type_ant, skills_enabled, and mcp_connected.

    Why it matters: The author breaks down the dynamic assembly logic behind Claude Code's system prompt, showing how conditional context engineering works in practice.

4/2Thu
  1. 陈与小金 · AI Coding 博客39

    从 Notion 重度用户到让 AI 帮我整理电脑：一位开发者的文件管理心得

    一位 Notion 重度用户转向用 Claude Code 等 AI Agent 协作后，放弃了花半个月搭建的复杂工作库，改为每月定期清理电脑文件。他建议只保留一个大类、按项目细分、用 Git 管理状态，并用 iCloud 加软连接映射 Obsidian 路径，避免 AI Agent 浪费上下文和 Token 寻找文件。他还建议用“归档”文件夹替代废纸篓，因为归档语义清晰，对 Agent 更友好。

    Awaiting translation

3/30Mon
3/28Sat
3/25Wed
3/24Tue
3/21Sat
3/18Wed
  1. Hacker News · AGENTS.md71

    把 AGENTS.md 当作目录而非说明书，附生成提示词

    作者认为多数 AGENTS.md 写得太长，把编码规范、架构决策、产品背景全塞进一个文件，而 LLM 每次启动任务都要读一遍，大部分内容与当前工作无关。他建议把 AGENTS.md 控制在约 100 行，只作为目录链接到 docs/ 下的架构、规范、测试等文档和产品背景文档，让智能体按任务自行拉取所需内容。

    Awaiting translation

3/17Tue
  1. 宝玉87

    How Anthropic’s Team Uses Claude Code Skills: Nine Categories and Writing Tips

    Thariq Shihipar, an engineer on Anthropic’s Claude Code team, summed up what the team learned from using hundreds of active Skills internally, sorting them into nine categories: library and API references, product validation, data acquisition and analysis, business process automation, code scaffolding, code quality and review, CI/CD and deployment, operations runbooks, and infrastructure operations.

    Why it matters: Anthropic’s internal classification system for hundreds of Skills, along with its writing tips, can be adapted to help teams design their own Skills.

  2. Paper Compute · Engineering Blog78

    日志即自愈反馈回路：用遥测让智能体跨会话积累经验

    作者让智能体在 stereOS 虚拟机里用 PyBoy 无头运行宝可梦红，速度约为实时的 100 倍，智能体自己输出 NAV、BATTLE、BACKTRACK 等日志前缀，这些日志经 tapes 代理流入 Kafka，再由 Flink SQL 做 STUCK_LOOP、TOKEN_SPIKE 异常检测，JSONL 与 DuckDB 负责跨会话查询。

    Awaiting translation

    Why it matters: 作者用终端里跑宝可梦的智能体做实验，展示日志如何变成跨会话的观测记忆并反哺下一轮运行。

3/16Mon
  2. 宝玉78

    The 8 Levels of Agent Engineering: From Tab Completion to Autonomous Agent Teams

    Bassim Eledath breaks the practical path of AI-assisted programming into 8 levels, from tab completion and agentic IDEs to context engineering, compound engineering, MCP and Skills, Harness Engineering, background agents, and finally autonomous agent teams.

    Why it matters: The author lays out AI-assisted programming as 8 levels, from tab completion to autonomous agent teams, so readers can figure out where their own team stands.

3/15Sun