今日已更新 100 条资讯 | 累计 24386 条内容
关于我们

标签：#pens

找到 1609 篇相关文章

AI 资讯

pip install self-audit: A Zero-Dependency CLI for AI Output Quality

AI agents pass tests while producing sloppy thinking. They say "should work" without evidence. They present partial work as complete. They embellish. I built a tiny tool that catches this. It checks any text across four dimensions: Completeness, Consistency, Groundedness, and Honesty. Install pip install self-audit Usage echo "Should work fine. Ready to ship." | self-audit --verbose Completeness: FIXED Groundedness: FIXED [should work fine] FAIL Zero dependencies. Python 3.8+. Stdlib only. 60 lines of core logic. The Four Dimensions Dimension Question What it catches Completeness Did I answer everything? Missing requirements Consistency Did I contradict myself? A-and-not-A patterns Groundedness Did I show evidence? "should work" claims Honesty Am I honest about limits? Embellishment, TODO stubs The dimensions are grounded in Anthropic Constitutional AI framework — Completeness (helpfulness), Groundedness (harmlessness), Honesty (truthfulness), Consistency (rule alignment). Try it on your own output After any AI-assisted coding session, pipe the agent text through self-audit before shipping. You will be surprised what it catches. GitHub: https://github.com/YuhaoLin2005/self-audit Claude Code skill: https://github.com/anthropics/skills/pull/1361

2026-06-27 原文 →
AI 资讯

I built a compiler for how AI agents should write to you

I kept correcting AI agents in the same ways: "too long," "answer first," "use a diagram," "assume I know the jargon." Each correction improved the current exchange, but the preference was not represented as durable system state. I built /calibrate-comms to make that state explicit. It is an open-source skill inside an Obsidian vault, used by both Claude Code and Codex. The model: nine operational dials The skill does not try to discover a personality type. It calibrates nine choices that directly change how an answer is rendered: Dial Practical question Density Tight sections or full reasoning? Sequence Answer first or chronological build-up? Modality Prose or diagrams for relational content? Abstraction Concrete example or principle first? Tradeoff One recommendation or several options? Detail Main path or edge cases too? Jargon Define terms or assume expertise? Tone Casual, neutral, or formal? Context-giving Should the agent extract missing context or split an overloaded brief? Prior → calibration → directives The workflow has three stages: L1 PRIOR → L2 SAMPLE REACTION → COMPILE → CLAUDE.md hypothesis empirical override shared by both agents Quick mode asks one bespoke forced-choice proxy per axis. Those questions are deliberately labelled as proxies, not validated psychometrics. Deep mode must fetch the exact items from supported open-access instruments before use; if the source cannot be obtained, the skill stays in Quick mode and makes no validation claim. The prior is then challenged through pairwise samples. For sequence, the contrast looks like this: Build-up first: We traced the latency spike to N+1 queries, then found lazy loading in a loop—so the fix is eager loading. Answer first: Fix: eager-load the association. Why: lazy loading in a loop caused N+1 queries and the latency spike. The user's pick is revealed preference. If it contradicts the prior, the sample wins. The compiler is the useful part The final profile is not a score report. A deterministi

2026-06-27 原文 →
AI 资讯

Day 6: my language now compiles to WebAssembly — and I emit the bytes by hand

I'm building LOOM — a small open-source language that is a machine-checked trust layer for AI-written code. I don't write it by hand anymore: an organism I built grows it, day and night, on my own machine. This is Day 6, and the whole day went to one thing — WebAssembly . Why this was a real test LOOM already runs three ways: an interpreter, and backends that compile checked code to Python and JavaScript. The thesis is "trust survives translation" — effects and provenance, proven once, hold the same on every target. WebAssembly is the strongest test of that: a low-level stack machine with linear memory, nothing like Python or JS. And there was a constraint. This machine's clang has no wasm target, and I install nothing paid or heavy. So I don't compile to wasm through a toolchain — I emit the wasm bytes myself (LEB128, the type / function / memory / global / export / code sections, the i32 stack machine) and run them through node's built-in WebAssembly . Zero dependencies. From fib to a value runtime, in a day Every step was prototyped and proven (wasm output == interpreter output) before it touched the kernel: The integer core — arithmetic, comparison, if , first-order calls and recursion. fib(10) becomes 61 bytes of real WebAssembly and returns 55, identically on the interpreter, Python, Node and wasm. A value runtime — let and integer lists in a real linear-memory heap (a bump pointer + a $cons cell allocator; head / tail are i32.load , empty is i32.eqz ). A list sums and folds by recursion, inside wasm. Sum types — (variant Tag e) becomes a tagged cell [tag-id | payload] ; match loads the tag, compares, binds the payload, branches. You can watch it: the live playground has a Compile → WAT button and WASM · fib / list-sum / match examples. Type a program, see it become real assembly, in your browser. Honest scope: ints, let , integer lists and sum types compile to wasm today. Records, closures and effects are the next frontiers (closures are the hard one — a func

2026-06-27 原文 →
AI 资讯

TMX: The open standard AI agent memory has been waiting for

TMX: The open standard AI agent memory has been waiting for The problem no one talks about: your agent's memories are prisoners. If you build an AI agent today using Mem0, your memories are locked in Mem0. Switch to Zep? You lose everything. Move to a new framework? Start from zero. This is exactly the problem email had in 1970. Every system had its own format. You couldn't send an email from one system to another. Then SMTP was invented. And email became universal. Today I'm publishing TMX v0.1 — the SMTP of AI agent memory. What is TMX? TMX (Truvem Memory eXchange) is an open, model-agnostic JSON format for storing, exporting, and importing AI agent memories across any platform, framework, or provider. It looks like this: { "tmx_version" : "0.1" , "exported_at" : "2026-06-26T20:00:00Z" , "source" : "truvem" , "agent_id" : "my-agent" , "memories" : [ { "id" : "550e8400-e29b-41d4-a716-446655440000" , "content" : "User prefers dark mode and concise responses" , "created_at" : "2026-06-01T08:30:00Z" , "updated_at" : "2026-06-01T08:30:00Z" , "expires_at" : null , "tags" : [ "preference" , "ui" ], "source_model" : "gpt-4o" , "metadata" : {} } ] } That's it. Plain JSON. Human-readable. Portable. Why this matters Right now, the AI agent ecosystem is exploding. Every week there's a new memory provider, a new framework, a new cloud service. But every one of them uses a proprietary format. This means: Developers are locked to their first choice forever Agent memories can't travel between clouds Switching providers = losing everything your agent learned This is the biggest hidden tax in the agentic AI stack. TMX fixes it with a single open spec that anyone can implement — for free, with no approval needed. The 5 core principles 1. Open — No license required. Implement TMX in any product, commercial or otherwise. 2. Model-agnostic — Works with GPT-4, Claude, Gemini, Mistral, Llama, or any future model. 3. Framework-agnostic — LangChain, CrewAI, Mastra, AutoGen — doesn't matter

2026-06-27 原文 →
AI 资讯

Back to Simplicity: Why We Built CALM, a Lock-Free Single-Thread Messaging Library for .NET

Hello everyone! For many years, I have been developing equipment control software and long-term support products using .NET/C#. Based on the experiences gained through working with various development teams, I would like to talk about why I created CALM (Cooperative Async Lock-free Messaging) , an open-source library for .NET. The Magic of UI Thread + Event-Driven + async/await My journey with .NET/C# began around the days of .NET Framework 4.0. At the time, code-behind in Windows Forms was incredibly intuitive. It allowed me to join the development team and become productive in a very short period. Back then, executing time-consuming operations on a separate thread and returning the results to the UI thread via callbacks was painful and error-prone. However, the introduction of async/await in .NET Framework 4.5 completely shifted the paradigm. The SynchronizationContext magically handled thread marshaling behind the scenes, and our code became amazingly simple. "Running asynchronous operations safely on a single thread (the UI thread) via an event-driven approach" — this seamless developer experience was my starting point. Divide and Conquer, Dependency Management, and CQS As our product evolved, the codebase grew, and the development team expanded beyond a certain size. Suddenly, the software became exponentially complex. Following industry best practices, we introduced MVP and MVVM patterns to separate the UI from the business logic. We also refactored our domain models based on Domain-Driven Design (DDD) and Clean Architecture principles, alignment with the team's domain knowledge. While this helped organize the logic within individual models, it introduced a new nightmare: complex dependencies between models. We struggled heavily with initialization, especially with models that had circular or mutual dependencies. To break this web of tight coupling, we introduced a mechanism that combined the Observer pattern (inspired by Android's EventBus) with the philosoph

2026-06-27 原文 →
AI 资讯

Asking vs Delegating AI Agents 🧐

Most developers use AI like a smarter Stack Overflow . Type a question. Get an answer. Go do the work yourself . That's fine but it's the slow way 😩 There's a faster mode, and most people haven't switched to it yet. Diff: Asking & Delegating When you ask an AI : "How do I write tests for my auth module?" You get a nice explanation. Then you write the tests yourself. You're still doing the work 🥸 When you delegate to an AI agent: "Write tests for /src/auth.py . Cover login, logout, and invalid token cases. Run them. If any fail, fix the code until they pass. Tell me what you changed." The agent opens your files, writes the tests, runs them, reads the failures, fixes the code, and comes back to you with a working test suite. You review the result. You didn't do the work. That's the shift 🙂‍↔️ It sounds small. The time difference is huge . How to write a good delegation Every delegation that works has four parts . Think of it like giving a task to a new team member: Goal: what should it produce? Scope: which files or area of the codebase? Success condition: how do we know it's done correctly? Report back: tell me what you changed and why. Here's what that looks like in practice: Debugging: "Here's the error and the stack trace. Find the root cause, fix it, and explain what was broken." Why this works: You're not asking what the error means. You're handing over the whole problem, find it, fix it, explain it 😎 Refactoring: "Refactor this file. Max two levels of nesting. No single function longer than 30 lines. Update every call site in the codebase." Why this works: The constraints are clear and checkable . The agent knows exactly when it's done 🧐 Database migration: "Write a migration script for this schema change. Make it idempotent. Run it against a local test database and confirm it succeeds." Why this works: You gave it a way to verify its own work before coming back to you 🤔 PR review: "Read this PR diff. Find anything that could fail in production. Write the tests

2026-06-26 原文 →
AI 资讯

Handling React Dialog Flows with async/await

React dialogs often start simple. You add an isOpen state, then a selected item state, then confirm/cancel callbacks, then another dialog after the first one. Eventually, a simple flow can become scattered across multiple components. For example, a user flow like this: Select a user Confirm the action Add the user often becomes multiple pieces of state: const [ isUserSearchOpen , setIsUserSearchOpen ] = useState ( false ); const [ isConfirmOpen , setIsConfirmOpen ] = useState ( false ); const [ selectedUser , setSelectedUser ] = useState < User | null > ( null ); This works, but the actual flow is harder to read. What if dialogs could be handled as async flows? I wanted the code to read closer to the user flow: const user = await openAsync ( UserSearchDialog ); if ( ! user ) return ; const confirmed = await openAsync ( ConfirmDialog , { title : `Add ${ user . name } ?` , }); if ( confirmed ) { await addUser ( user . id ); } The dialog opens, waits for a result, and the caller continues based on that result. This is about orchestration, not UI This is not meant to replace Radix, MUI, Headless UI, shadcn/ui, or custom dialog components. Those libraries solve the dialog UI problem well. The idea here is to manage the flow around dialogs: opening dialogs from anywhere under a provider resolving typed result values handling nested dialogs distinguishing completed vs dismissed supporting dismissal reasons guarding close behavior with shouldClose So the actual dialog UI can still be your own component. I packaged the pattern I turned this idea into a small open-source library called react-dialog-flow . It provides a headless dialog stack, Promise-based openAsync , typed results, nested dialogs, closeTop , closeAll , dismissal reasons, shouldClose , and optional UI primitives. GitHub: https://github.com/CHOKANGYEOL/react-dialog-flow npm: https://www.npmjs.com/package/react-dialog-flow Docs: https://dialog-flow.kangyeol.com/ It is still early, so I am mainly looking for feed

2026-06-26 原文 →
AI 资讯

Dotsync – A zero-knowledge CLI to sync dotfiles and secrets in Go

I spent 3 months building a tool that solves the most annoying part of working on a dev team. You know the drill: → New dev joins → "hey check your DMs" → Someone pastes the .env in Slack → That message sits there forever → 6 months later someone screenshots it by accident or worse — → git add . → git commit -m "added env" → git push → you just leaked your prod database URL to the internet I got tired of it. So I built DotSync. ─────────────────────────────────── Here's what it looks like in practice: $ dotsync push 🔒 Encrypting 10 secrets for team access... 📤 Uploading... ✅ Version : v7 Secrets : 10 keys encrypted Teammates can now run: dotsync pull That's it. New dev joins? They run dotsync pull. Secrets updated? dotsync push. Moving between your laptop and work machine? dotsync pull. ─────────────────────────────────── The part I actually care about — security: Everything is encrypted ON your machine before it hits the network. The server stores a blob it literally cannot read. I'm not asking you to trust my infrastructure. You don't have to. Stack if you're curious: • Argon2id key derivation (64MB memory cost — brute force isn't happening) • AES-256-GCM encryption • Zero-knowledge server (stores ciphertext only) • Single Go binary, no runtime deps ─────────────────────────────────── Other things it does that I use daily: dotsync diff → shows exactly which keys changed vs remote (never shows values) dotsync history → full version history, who pushed what and when dotsync rollback → restore any previous version in one command dotsync run -- node server.js → injects secrets as env vars, nothing written to disk dotsync scan → scans your codebase for accidentally committed secrets ─────────────────────────────────── Free tier covers: • 1 project • 3 team members • 7 days history That handles most small teams completely free. ─────────────────────────────────── Now the part where I need your help: I'm looking for 100 engineers to actually stress test this. Not "give i

2026-06-26 原文 →
← 上一页 下一页 →