今日已更新 253 条资讯 | 累计 41930 条内容
关于我们

标签：#open

找到 2922 篇相关文章

AI 资讯

Five ways your LLM cost tracking is lying to you

Your monthly OpenAI or Anthropic invoice tells you how much you spent. It doesn't tell you which feature spent it, which model, or why last Tuesday cost three times as much as Monday. So at some point you (or your team) will build a metering layer: wrap the client, read usage off the response, multiply by a price table, ship it to a database. I did exactly that over the past few months while building an LLM observability service, and my numbers were wrong in five different ways before they were right. Every one of these failures was silent. No exception, no alert, just numbers that were quietly too low or too high. This is the list I wish someone had handed me. Pitfall 1: Streaming responses quietly report zero tokens OpenAI's Chat Completions API returns no usage data at all for streaming requests unless you pass stream_options: { include_usage: true } . No error, no warning. The stream just never contains token counts. If your metering reads usage off the chunks, every streaming call gets recorded as 0 tokens, $0. And since chat UIs are almost always streaming, that's most of your traffic. This one bit me twice in the same audit. First finding: all streaming calls in my own dashboard were $0. Second, nastier finding: I had a budget-gate feature that blocks calls once spend crosses a limit, and it waved every streaming call straight through — because as far as it could tell, streaming was free. The fix is to inject the option in your wrapper when the caller didn't set it: let injected = false ; if ( params . stream && params . stream_options ?. include_usage === undefined ) { params = { ... params , stream_options : { ... params . stream_options , include_usage : true }, }; injected = true ; } But there's a trap inside the fix. With include_usage on, OpenAI appends one extra chunk at the end of the stream that carries usage and has an empty choices array . Any downstream code that does chunk.choices[0].delta — which is most example code on the internet — will throw

2026-07-13 原文 →
AI 资讯

Why Your Team's AI Assistant Acts Like It's the First Day on the Job, Every Single Time

Anyone who has used AI tools for a while has probably run into this annoyance. You ask it to write a weekly report in the morning and it doesn't know your KPI framework was overhauled last week. You ask for a technical proposal in the afternoon and it has no idea you spent three months locking down your tech stack. Every new conversation means re-explaining the project background, which decisions were made and why. In multi-person collaboration the problem scales up fast. Five people each interacting with AI separately; the AI's understanding of each person is isolated. A discusses an architecture decision with the AI, B has no idea that conversation happened. Five people are repeating the same explanations and none of them know the others already did. Context Fragmentation Has Nothing to Do with Model Capability Current mainstream AI tools store memory as conversation history stuffed into a context window. When the window fills up, older messages get truncated. That works fine for a single conversation but falls apart in cross-day, cross-week team collaboration. Even with 128K token support, cramming all project history in there causes information density to collapse and the model loses the ability to focus on what matters. Team collaboration needs memory across several layers. Project background, tech stack choices, the reasons behind past pivots; this long-term context doesn't appear in any single conversation but affects every task. One team member prefers concise communication while another wants detailed reasoning; the AI should remember these differences instead of outputting the same format for everyone. Last week's design decision and why it went that way, how that choice affects this week's sprint planning; if the AI can't see these connections, its suggestions will clash with earlier direction. Some products use vector retrieval to extend memory, storing past conversations as embeddings and recalling relevant snippets by semantic similarity when needed. T

2026-07-13 原文 →
AI 资讯

Why I Built an Adversarial Co-Generation Engine

I spent a chunk of last year around legacy modernization work — the kind of project where a bank or an insurer is taking twenty years of accumulated code and rebuilding it as modern services, one system at a time. Every one of those systems starts the same way: a PRD or a requirements document says what the business needs, that gets translated into a spec precise enough for an AI to implement, and eventually someone tests what came out. What struck me, watching this happen at scale, wasn't that the code was bad. It was that nobody was testing the thing that actually determined whether the code would be bad: the spec itself — the technical description handed to the model, not the PRD that motivated it. Every security tool I looked at — SAST scanners, DAST tools, even the AI coding assistants themselves — waited until an implementation existed before doing anything adversarial. Attack the code, once it's there. That's the whole industry's model, and it's worked fine for forty years because the volume was always survivable. A team ships a handful of PRs a week, a human reviews them, and eventually a pentest catches whatever slipped through. That math falls apart at modernization scale. When you're regenerating a few million lines of code, you're also generating a few thousand specs, faster than any review process was ever built to absorb. Testing after the fact doesn't just get slower under that load — it quietly stops happening, spec by spec, until the aggregate exposure is enormous and nobody can point to when it happened. So I built GAUNTLEX to test the thing that happens before the code does: the spec. This is also where I want to be precise about a word that gets overloaded. "Spec-driven development" — the broader industry shift toward writing structured, agent-facing specs instead of prompting an AI free-form — is exactly the world GAUNTLEX lives in. But a spec (what to build, precise enough for a model to implement) and a PRD or requirements doc (why it's needed

2026-07-13 原文 →
AI 资讯

FROST周报 | 为什么智能体需要「谱系」？从生物学隐喻看AI治理新范式

FROST周报 | 为什么智能体需要「谱系」？从生物学隐喻看AI治理新范式 作者按 ：本文是 FROST 开源项目的每日推广系列文章，周一深度篇。 一、一个被忽视的根本问题 当我们谈论 AI Agent 时，大多数讨论都聚焦于「能力」：能不能写代码？能不能调用工具？能不能规划任务？ 但有一个根本问题很少被触及： 当一个 Agent 执行了错误的决策时，谁来负责？当它消亡后，它的经验能否被传承？ 就像一个没有记忆的人，每次醒来都是白纸一张——这不叫智能体，这叫复读机。 FROST 正是为了解决这个「治理真空」而诞生的。 二、从细胞分裂到 Agent 家族 FROST 的核心哲学只有一句话： 细胞会死，但谱系会存续。Agent 会消亡，但宪法会传承。资产会永存。 这不是文学修辞，而是一套完整的技术架构。 四个原子：最小可行集合 FROST 只定义了四个原子，却能构建任意复杂度的智能体系统： 原子 职责 生物类比 Store 记忆容器，只做 save/load/delete 细胞核 Skill 纯能力单元，无状态无副作用 蛋白质 Agent 膜包裹的细胞，拥有 Store + Skills 神经细胞 SOP 有序步骤列表，可教学、校验、优化 宪法文本 from core import Store , Agent , skill_set , skill_get # 创建一个最小 Agent store = Store () agent = Agent ( " cell " , store , skills = { " set_context " : skill_set , " get_context " : skill_get }) # 执行任务 result = agent . run ( sop_steps = [ " set_context " , " get_context " ], initial_context = { " key " : " message " , " value " : " FROST is alive " } ) # result["_result"] == "FROST is alive" 关键洞察 ：Store、Skill、Agent、SOP 这四个概念彼此正交，可以自由组合。就像乐高积木，从简单到复杂，始终保持可解释性。 三、家族治理：超越扁平架构 传统的多 Agent 系统通常是扁平的：所有 Agent 平等对话，没有层级，没有记忆，没有责任边界。 FROST 引入了「家族治理模型」——一个三层递归结构： 祖辈 (Ancestor) ：定义不可违背的宪法与长期目标 父辈 (Parent) ：领域协调者，可递归委托 孙辈 (Leaf) ：执行具体原子任务，瞬态存在 四个协议保障治理闭环 ： 层级 Store 继承 ：祖先记忆只读，后代自动继承 SOP 宪法校验 ：祖辈审核后代 SOP，拒绝违规执行 编排层级限制 ： max_spawn_generation 硬编码，禁止越级 spawn 选择性持久化 ：父辈收割有价值产出，淘汰冗余 Agent 四、V5.0 五维元模型：多维治理架构 2026年7月发布的 V5.0 引入了一个重大升级—— 五维元模型 ： 维度 模块 核心职责 武器注册表 Armory 能力的元数据管理与发现 任务注册表 TaskRegistry DAG 任务编排与图谱 SOP 事件编目 EventCatalog + Strategist 态势感知与双模式事件分析 平台注册表 PlatformRegistry 外部能力的发现、调用与健康检查 规则注册表 RuleRegistry 可版本化的治理约束与合规检查 197 个测试用例 保障了每个维度的质量。 五、与现有框架的差异 维度 LangChain CrewAI FROST 状态管理 链式传递 角色记忆 层级 Store 权限边界 无 提示词软约束 代码强制只读 治理可审计 无 对话日志 结构化执行历史 架构无关 ✅ ✅ ✅ FROST 不重复造轮子。它填补的是「治理」这个空白地带： 让多智能体系统真正可控制、可追溯、可进化 。 六、快速体验 # 克隆仓库 git clone https://gitee.com/liao_liang_7514/frost.git cd frost # 运行测试 python -m pytest # 查看示例 python frost_run.py 完整文档： https://gitee.com/liao_liang_7514/frost 七、写在最后 AI Agent 的下一阶段，不是更强的模型，而是 更好的治理 。 当我们把 100 个 Agent 放在一起时，如果没有宪法、没有层级、没有记忆传承

2026-07-13 原文 →
AI 资讯

Dev log #12 Hardening WebRTC and Polishing the UI: A Week of Networking and Refinement

Spent the week balancing deep p2p networking work in Python with some much-needed UI polish on my personal site. 11 commits and 6 PRs later, I hit a perfect 7-day streak and made the codebase a bit more secure. TL;DR This week was all about the "invisible" work that makes software feel solid. I spent a good chunk of time in the weeds of p2p networking, specifically hardening WebRTC implementations, while also carving out time to refine the typography and feel of my personal portfolio. With 11 commits across 5 repos and 6 PRs in flight, I managed to keep the momentum going every single day of the week. WHAT I BUILT Most of my direct commit activity this week was split between keeping my dev environment sharp and making my portfolio feel a bit more "me." Portfolio & Personal Branding I spent some quality time in yashksaini-coder/portfolio . If you're like me, you can't leave your personal site alone for more than a month. I pushed a few updates to the blog content, but the real fun was in the UI/UX tweaks. I swapped out the primary typography for JetBrains Mono —there’s just something about a good monospace font that makes a dev portfolio feel right. I also went through a "make-interfaces-feel-better" phase. I refactored the selectedwork section, specifically dropping a cursor-follow preview tile that felt a bit too "heavy" and replaced it with something more streamlined. I also polished the index rows to make the transitions feel snappier. It’s about +452/-279 lines of code, which is a healthy amount of churn for a week that was supposed to be about "minor" updates. The Maintenance Grind My nvim config is basically a living organism at this point. I have CI set up to automatically track plugin updates, and this week was particularly noisy with 6 commits just keeping the toolchain current. It’s [skip ci] territory, but it ensures that when I sit down to actually write code, my editor isn't lagging behind the latest Lua API changes. I also did a quick version bump for

2026-07-13 原文 →
AI 资讯

Blocking AI crawlers earns you nothing. Here's how to price them instead

Disallow: GPTBot is a wall. Walls don't pay rent, and the crawlers that matter most either ignore them or route around them. If your content is worth training on, the interesting question isn't "how do I keep the bots out" — it's "what do they owe me, and how do I say so in a way a machine can read." That's what RSL (Really Simple Licensing) is for. It shipped 1.0 in December 2025 with around 1,500 publishers behind it — Reddit, Yahoo, Quora, O'Reilly, Medium, Vox. This post is a from-scratch walkthrough of what the format actually is, the six places you can put it, the one mistake that makes crawlers silently ignore your terms, and where the declaration stops and enforcement begins. No tooling required to follow along — it's all plain XML and HTTP. The format is an XML vocabulary, not a config file An RSL document says: for this content, here's what's permitted, what's prohibited, and what it costs. Minimal example: <?xml version="1.0" encoding="UTF-8"?> <rsl xmlns= "https://rslstandard.org/rsl" max-age= "7" > <content url= "/" > <license> <permits type= "usage" > search </permits> <prohibits type= "usage" > ai-train </prohibits> <payment type= "crawl" > <amount currency= "USD" > 0.015 </amount> </payment> </license> </content> </rsl> Read it out loud: search engines may index this; training on it is prohibited; if you want to crawl it anyway, the rate is $0.015. usage tokens include search , ai-train , ai-use (inference/grounding), and a few more. You can scope rules by user and geo too. One rule that trips people up: prohibition wins . If the same token shows up under both permits and prohibits , the content is prohibited. Don't try to express "allowed except for X" by listing X in both — just prohibit X. The namespace is the thing crawlers actually key on The single most common way to publish RSL that quietly does nothing: getting the namespace wrong. It must be exactly: xmlns="https://rslstandard.org/rsl" http instead of https , a trailing slash, or a plausible

2026-07-12 原文 →
← 上一页 下一页 →