精选归档 · 第 2 页

第 21–40 条 · 共 694 条

9月10日9月10日周四

星期四 · 1 条

00:00

Cognition 模型 / Devin 博客（网页）精选

AI 评分 70/100

Cognition 发布 SWE-2 编码模型，以更低成本逼近前沿

Cognition 发布最先进编码模型 SWE-2，基于 2.8T 参数的 Kimi K33 后训练，在 FrontierCode 1.1 Main 取得 50.0%，仅比 Fable 5.1 低不到一分且便宜 64%。

推理数据/训练模型发布编码

另有 6 家信源报道

推荐理由：官方详细公开了 SWE-2 的训练方法与成本惩罚公式，读者可以借此理解如何用 RL 推移整条成本性能前沿。

9月9日9月9日周三

星期三 · 5 条

08:20

公众号：数字生命卡兹克精选

AI 评分 70/100

GPT-6 Astra推理等级怎么选才最省Token

卡兹克发文讲解GPT-6 Astra的推理强度等级（Reasoning Effort）含义，指出各档位是同一模型的不同思考预算，Ultra则类似拉起多个智能体协作的专项工作组。

OpenAI 推理教程/实践编码

推荐理由：作者基于烧完两个200刀会员的第一手使用体感，给出推理等级的通俗解释和分档位省Token的具体用法。

08:00

Together AI 研究与产品博客（RSS）精选

AI 评分 67/100

Together AI 深度解析开源模型 AI 编码栈 MIGHT

Together AI 发布深度解析文章，介绍开发者从闭源模型转向开源模型所需的 AI 编码栈，提出由 Model、Inference、Gateways、Harness、Tools 组成的 MIGHT 框架。

智能体 MCP/工具开源生态教程/实践

推荐理由：原文给出开放权重模型编码栈的分层框架，读者可据此按任务选模型、按价格选供应商并保持工作流稳定。

01:59

Claude：Blog（网页）精选

AI 评分 66/100

Anthropic 发布 Claude Platform 降本指南并更新 claude-api 技能

Anthropic 介绍用 Claude Platform 降低成本、保持性能的三种方法：提高提示词缓存命中率、升级到前沿模型时清除提示词反模式、按任务校准 effort。

教程/实践编码部署/工程

推荐理由：官方给出了提示词缓存、反模式清理和 effort 校准三类降本方法及实测数字，可直接迁移到自己的 Claude API 应用。

01:07

ClaudeDevs@ClaudeDevs精选

AI 评分 67/100

Anthropic 讲解用 Claude Platform 降低成本并提升性能的三个方法https://x.com/i/article/2095208140753317888Reducing cost and improving performance with Claude PlatformTuning prompt caching, instructions, and effort can reduce Claude's cost without sacrificing application performance.Performance and cost are often viewed as a trade-off: to spend less, you accept worse results. In practice, we've found that many applications using Claude Platform can cut cost without giving up performance with three fixes: maximize the prompt cache hit rate, remove anti-patterns from your prompts when upgrading to frontier Claude models, and calibrate effort to the task. We've put this guidance into the claude-api skill. In this article, we show how Claude Code with the claude-api skill can often find ways to reduce cost while maintaining or improving performance.Prompt cacheBefore Claude generates a response, it first processes your prompt into an internal working state. This step, called prefill, is the expensive part of handling input. Prompt caching saves that state (the key–value, or KV, cache): when a request starts with the same prefix, Claude reads it back instead of recomputing it. Cache reads are billed at a fraction of the full input price.There are a few practical considerations to ensure effective use of the prompt cache. First, the prompt cache is pinned to a specific model. Second, prompt cache reads must be byte-exact across the prefix. Finally, the prompt cache has a limited time-to-live (TTL).With these points in mind, there are a few practical tips:• Be careful changing effort setting mid-conversation. These settings render into the prompt ahead of your content, so they are part of the cached prefix. Only with select Claude models, including Opus 5 and Fable 5.1, can you update effort mid-conversation without breaking the cache.• Keep volatile values out of the prefix. A dynamic timestamp or ID in the system prompt can change across model calls, and break the cache.• Avoid tool definitions that reorder themselves. When using the Claude messages API, the prompt is assembled in a fixed order with tool definitions rendered at the top. Any change to the tool definition will break the cache.• Be careful when forking conversations. Subagents and branches only share the parent’s cache when the fork’s prefix is byte-identical, on the same model, and using the same effort.How to fix itWe’ve accumulated a few lessons for prompt cache management:• Monitor your prompt cache hit rate carefully. Claude Console and the cache diagnostics API provide prompt cache diagnostics, including reasons for prompt cache misses (Figure 1) and exactly where two requests diverged.• Defer rarely used tools. Declare all your tools up front but mark the rarely used ones defer_loading: they stay out of the cached prefix and are appended into the conversation only when Claude looks them up with tool search, so the cache is preserved.• Apply system prompt updates as messages. Certain Claude models let you add a system instruction as a message mid-conversation instead of editing the system prompt, which preserves the cache.• Lay out the request so the stable part stays stable. Add static context (tool definitions and the system prompt) first and the growing conversation behind them (Figure 2).• Make changes to model or effort when the prompt cache will already be broken. Certain operations, like compaction, already rewrite much of the cache (the conversation). That is a good moment to switch model or effort, since you are paying for a miss anyway.• Move the cache breakpoint as the conversation grows. With Claude Platform, you can set automatic caching to apply the cache breakpoint to the last cacheable block.• Pre-warm the cache. To reduce latency, send a request with max_tokens: 0 and an explicit cache breakpoint, using the same effort setting as your real traffic. This processes the prompt and writes it to the cache without generating anything. If you run it at session start (for example, while a user is typing), the first real request hits a warm cache.• Don’t exceed the prompt cache TTL. The 5-minute cache TTL counts from the start of the request. If an agent blocks on tool calls or subagent requests that run longer than 5 minutes, the parent's cache expires before the result comes back. In cases like this, consider setting a 1-hour TTL on the prefix instead.InstructionsPrompts can accumulate instructions that patch model weaknesses. These instructions can drift relative to the capabilities of the latest Claude models. Here are common prompting “anti-patterns” that hobble frontier Claude models and can inadvertently increase costs:• Verification rituals. Instructions like "double-check your work” or "verify twice before responding” are often taken literally by frontier models and can waste tokens.• Thoroughness and emphasis boosters. "Be maximally thorough," "CRITICAL: YOU MUST ALWAYS…" can lead to verbosity and extra tool calls when working with frontier models.• Mandatory procedures and scratchpad scaffolds. Fixed-step processes (e.g., "think step by step in a scratchpad") or reasoning templates are rituals that frontier models don't need. This scaffolding can stack on top of native reasoning and use unnecessary tokens.• Stale examples. Few-shot examples tuned to an older model's failure modes can teach a frontier model to imitate long reasoning chains on requests that don't need them.• Contradictory rules. Frontier models are better at instruction following. Contradictory instructions ("always refund within policy" vs. "never issue refunds without escalation") can be followed more literally by frontier models, resulting in degraded performance.• Dated configuration. Settings written for an older Claude generation (e.g., manual thinking budgets) can be rejected by Claude Platform with newer models.How to fix itWe've updated the claude-api skill with a new command that watches out for these anti-patterns. In Claude Code, run /claude-api prompt-audit against your prompts, skills, or tool descriptions. The audit covers anything in your working directory, including application code that calls the Claude API and Claude Code's own configuration (e.g., CLAUDE.md or skills).For example, we tested a model migration from Opus 4.8 to Opus 5 on a customer support benchmark. We started from a clean prompt and planted one anti-pattern at a time (a retired thinking setting, a pair of contradictory refund rules, a manual scratchpad, "verify twice", "be maximally thorough", and a mandatory six-step procedure), giving six legacy prompts.We ran each on Opus 4.8, on Opus 5 with only the model ID changed, and on Opus 5 after running /claude-api prompt-audit once per prompt (Figure 3 shows the average across the six).With Opus 5, verification rituals ("verify twice") use unnecessary tokens by duplicating order lookup on every refund. Emphasis boosters ("be maximally thorough") became dozens of unneeded knowledge-base searches.Running /claude-api prompt-audit removed the anti-patterns, decreasing costs by 14.6% and increasing accuracy by 5.3% on average. Cost dropped because extra tool calls and duplicated reasoning were eliminated. Accuracy rose for three reasons. The retired thinking setting made the API reject every routing request outright. The contradictory refund rules led Opus 5 to withhold four refunds it owed while it asked the customer to confirm. And the manual scratchpad collided with Opus 5's built-in thinking: on three tickets it wrote the tool call inside its reasoning and never executed it.EffortEffort tells Claude “how hard to work.” At low effort Claude generally reaches conclusions faster. At high effort, Claude deliberates, verifies, and explores alternatives before answering.Cost-versus-performance across effort levels on a single model can vary. For example, Claude Fable 5 scores 11.5% at low effort for $5.35 per task on FrontierCode Diamond (the hardest 50 tasks). At max effort, Fable 5 gets 30.9% for $19.00 per task; changing effort raises the score about 2.7x (+19 points) for about 3.5x the cost (Figure 4).On Claude Fable 5.1, Humanity's Last Exam (without tools) shows a steep curve with a diminishing last step. It scores about 53% at low effort for about $0.30 per question and about 61% at max effort for about $2.23; the last step up to max adds about half a point for 46% more cost. The gain falls inside the benchmark's run-to-run noise, so you pay more for no measurable gain.Effort can be miscalibrated in either direction:• Assuming higher is always better. High effort can cause over-thinking. Claude spends more time deliberating than the task warrants, which adds cost and latency, and can degrade answer quality. Deliberation only helps while there's still evidence to find.• Biasing to low effort. Set too low, Claude stops before it has enough evidence. It makes fewer tool calls, so it may answer from the first search result instead of the third. It thinks less on hard steps and skips the check it would normally run on its own. The answer looks finished, but it's built on partial information.How to fix itThere are some useful ways to calibrate effort:• Test stronger models at lower effort. A stronger model at low effort can be cheaper than a weaker model working hard (high effort). For example, on CursorBench 3.2, Claude Fable 5.1 at low effort matches the performance of Fable 5 at high effort at a third of the cost (Figure 5). Two things make the newer model cheaper: at low effort it does less work per task, and Fable 5.1's prompt-cache reads are priced at $0.25 per million tokens versus $1.00 for Fable 5. Even at Fable 5's prices, Fable 5.1 at low effort would cost about 40% less.• Understand your task shape. Measuring application performance across a sweep of effort levels is a useful way to understand the cost-performance tradeoff for your particular task. On a non-saturated evaluation, a flat cost-performance curve across effort levels suggests that the task is not bound by thinking compute; increasing effort is not beneficial.This calibration often involves running an evaluation across models and effort levels. In Claude Code, /claude-api hillclimb performs this search for you: it splits your evaluation into train and test sets, proposes configuration changes, and reads failing train examples to fix what it finds.We ran it on a customer support benchmark, starting from Opus 4.8 at its default (high) effort. The hillclimber first tried Opus 5 at low effort, applying prompt-audit to remove mandatory tool-call rituals, scratchpad steps, and contradictory rules. That cleared the Opus 4.8 baseline at 98.9% train accuracy and cut cost to 2.6 cents per ticket (Figure 6).It then stepped down to Sonnet 5 at low effort, which was cheaper still at 1 cent per ticket, but accuracy fell to 88.9%. Reading the failing train tickets, Claude added routing rules and a refund-cap cross-reference to the prompt, bringing Sonnet 5 back to 98.9% at the same cost.On the 14 held-out tickets the search never saw, the final configuration scored 90.5% against the original setup's 78.6%, at about one-fifth the cost.Automating cost reductionPrompt caching, instructions, and effort are common levers for reducing cost. Our documentation covers even more. To run a holistic cost audit of application code that uses the Claude API, we've added /claude-api cost-optimize: it profiles where your spend goes, applies cost reductions, and, if you provide an evaluation, shows how savings trade off with performance.cost-optimize starts by finding where your tokens go: from your organization's usage and cost reports if you have a Claude Admin API key, from the usage object on each API response if your application logs it, or, failing both, by reading your request-building code and estimating.It then ranks the available savings, starting with prompt caching, trimming what each request carries (including a prompt-audit), bounding output, and batching unattended work. If you supply an evaluation, it goes further and computes cost and performance across effort levels and model choices. We ran this on four public benchmarks with Sonnet 5 as a baseline (Figure 7):• LegalBench (~58% lower cost): cost-optimize proposed caching a shared prefix across tasks, setting low effort, and processing tasks via the Batch API. Thinking tokens fell from 102,779 to 8,284 and pass rate stayed within noise and cost dropped by ~58%.• tau2-bench retail (~73% lower cost): By implementing prompt caching with explicit breakpoint placement, cost-optimize reduced spend by 72% while keeping pass rate flat.• OfficeQA Pro (~52% lower cost): cost-optimize added batch processing and document caching, which brought cost down from $136.20 to $64.87.• SWE-bench Verified (~55% lower cost): cost-optimize found that the default config already caches correctly. Savings came from setting effort to medium and constraining the agent’s output to just a few concise sentences. Median steps per task went from 29 to 17 and prompt tokens fell from 75.2M to 33.7M.Getting startedStart with /claude-api prompt-audit when you've migrated to a frontier Claude model and want to check your existing prompts. It scans the prompts, skills, and tool descriptions in your working directory. This can be application code that calls the Claude API or Claude Code's configuration (CLAUDE.md, skills). It removes common anti-patterns that hobble frontier models.Reach for /claude-api cost-optimize when your application uses the Claude API and you want a cost audit. It profiles token spend and then tests different levers: it applies prompt-audit, but also checks for ways to lower cost via prompt caching, batching unattended work, or bounding output. If you provide an evaluation, it measures the effort and model selection trade-offs.Finally, use /claude-api hillclimb for a search over cost and performance. Given an evaluation, Claude splits it into train and test sets, then proposes updates to your application that aim to reduce cost while maintaining baseline performance. Claude reads the failing train cases to guide the search, and the final configuration is scored on the held-out test set.To learn more:• See our documentation here• See our cost reduction cookbook here• See the claude-api skill here; the skill is also built into Claude Code• See this article on the Claude Blog hereWritten by Lance Martin (@RLanceMartin), Brad Abrams (@brada), Isabella He (@IsabellaKHe), and Ben Lehrburger (@benlehrburger).译Anthropic 团队文章指出，优化 prompt cache 命中率、清除升级到前沿 Claude 模型后的提示词反模式、校准 effort 三个手段可在不牺牲性能的情况下降低成本。

智能体 Anthropic 教程/实践编码

推荐理由：官方团队给出提示词缓存、反模式清理和 effort 校准三条降本路径，并用公开基准实测数字支持，方法可迁移到自己的 Claude 应用。

00:00

Cognition 模型 / Devin 博客（网页）精选

AI 评分 76/100

Cognition 团队用 Devin 完成 RSA-260 分解，刷新公开 RSA 挑战纪录

Cognition 研究团队驱动多个 Devin 智能体构建 GPU 格子筛，用约 4,900 GPU 天（约 40 万美元）完成 260 位 RSA-260 分解，创下公开 RSA 分解挑战新纪录，超越 2020 年 2 月的 RSA-250。

智能体编码论文/研究

另有 1 家信源报道

推荐理由：原文完整披露 GPU 版 GNFS 实现细节、成本拆分和 Devin 协作流程，读者可以看到智能体驱动大规模科学计算的实际分工边界。

9月6日9月6日周日

星期日 · 2 条

23:59

OpenAI：官网动态（RSS · 排除企业/客户案例）精选

AI 评分 81/100

OpenAI 发布内部研究加速报告：已达成自动化研究实习生目标，推进 2028 年 3 月自动化 AI 研究员

OpenAI 发文披露自动化研究进展，宣布已达成去年秋天设定的今年 9 月拥有自动化研究实习生（可在人类指导下完成耗时数天的明确研究任务）的目标，并计划在 2028 年 3 月前造出自动化 AI 研究员。

智能体推理现象/趋势编码

推荐理由：OpenAI 以内部数据披露 coding agent 对研究工作的实际影响和 RSI 进展，读者可以据此了解前沿实验室的自动化研究现状。

05:59

🚨 AI News | TestingCatalog@testingcatalog精选

AI 评分 76/100

OpenAI GPT-6 Astra 登顶 Code Arena WebDev 榜首，领先 Claude Fable 5.1 达 35 分GPT-6 Astra from OpenAI claimed a top spot on WebDev Arena, surpassing recently released Claude Fable 5.1 by 35 points.“It also reshapes the Pareto frontier as the best-performing model at $40/Mtoken, which matches the latest Claude model pricing.”Are we set to hit 2k by the end of this year?译GPT-6 Astra （Max）以 1797 分登顶 Code Arena： WebDev，领先第 2 名 Claude Fable 5.1 （Max） 35 分、第 3 名 Claude Opus 5 （Max） 1688 分。

Arena.ai: 真实世界的结果已经出炉。Code Arena 上出现了新的第一名--GPT-6 Astra （Max）！它还重塑了帕累托前沿，成为 $40/Mtoken 价位上性能最强的模型，这与最新的 Claude 模型定价持平。 OpenAI 的 G...

OpenAI 模型发布编码

推荐理由：榜单数据给出了与 Claude 系列的具体分差和同价位对比，读者可以据此评估新模型的实际编码位置。

9月5日9月5日周六

星期六 · 5 条

21:33

The Decoder：AI News（RSS）精选

AI 评分 78/100

OpenAI 发布 GPT-6 Astra 提示词指南，含 slop 词屏蔽清单

OpenAI 在模型文档中说明 GPT-6 Astra 相比 GPT-5.6 Sol 更常提出澄清问题、对上下文更敏感，并给出让模型更主动、审计 AGENTS.md 等技能文件、控制写作风格、约束子智能体委派和测试规模的提示词建议。

智能体 OpenAI 教程/实践编码

推荐理由：原文汇总了 OpenAI 官方文档中针对 GPT-6 Astra 的提示词建议和 slop 词屏蔽清单，开发者可直接迁移到自己的提示词写法。

19:40

公众号：数字生命卡兹克精选

AI 评分 77/100

实测GPT-6 Astra：速度、前端与代码能力对比GPT-5.6 Sol的全面升级

GPT-6 Astra正式向所有订阅用户推送，作者实测后认为其综合能力追平Claude Fable 5，且额度100%可用。相比GPT-5.6 Sol，速度明显提升，大型系统审查从数小时缩短到约10分钟，代码扫描找出大量此前未发现的性能问题并2小时完成修复；前端3D生成和审美大幅强化，写作在白描和用词上更好但仍缺中文留白感。

智能体 OpenAI 编码评测/基准

推荐理由：作者实测了GPT-6 Astra在速度、前端生成、代码深度和写作上的具体变化，并给出可迁移的AGENT.md简化思路。

07:07

Sam Altman@sama精选

AI 评分 79/100

GPT-6 Astra 开始向 Plus 和 Business 用户推出Now out to all Plus and Business users.Happy building!译Sam Altman 宣布 GPT-6 Astra 现已向所有 Plus 和 Business 用户推出。此前该模型已面向 Pro、Enterprise 和 Business Premium 用户在 Work/Codex 及 API 中提供。

Sam Altman: GPT-6 Astra 现已面向 Work/Codex 中的所有 Pro、Enterprise 和 Business Premium 用户开放，并已在 API 中提供。我们接下来将开始向 Plus 和 Business 用户推送。感谢大...

OpenAI 推理模型发布编码

推荐理由：原文确认 GPT-6 Astra 的用户覆盖范围扩大到 Plus 和 Business，读者可以据此判断自己的可用入口和时间点。

03:59

OpenAI@OpenAI精选

AI 评分 81/100

OpenAI 发布 GPT-6 Astra，面向 Pro、Enterprise 和 Business Premium 用户开放GPT-6 Astra is now available to all Pro, Enterprise, and Business Premium users in ChatGPT Work and Codex. It's also live in the API.It might take a few days to roll out to our Plus and Business users. Thank you for your patience.译OpenAI 宣布 GPT-6 Astra 现已向所有 Pro、Enterprise 和 Business Premium 用户开放，可在 ChatGPT Work 和 Codex 中使用，同时已上线 API。Plus 和 Business 用户的推送可能需要几天时间。

OpenAI 模型发布编码

推荐理由：官方宣布 GPT-6 Astra 上线范围与渠道，Plus 和 Business 用户还需等待几天，读者可据此确认自己能否用上。

00:29

GitHub Blog精选

AI 评分 68/100

GitHub 发布 Project HydraFusion 研究预览，用多模型运行时编排降低 Copilot 成本

GitHub 推出 Project HydraFusion 研究预览，通过运行时多模型编排，在 Single、Cascade、Critique 三种执行模式间为每个任务选择工作流，以平衡质量、成本和延迟。

智能体 GitHub 产品更新编码

推荐理由：原文给出三种执行模式和三个基准的质量与成本数据，读者可以据此评估多模型编排对编码工作流的影响。

9月4日9月4日周五

星期五 · 3 条

08:32

Hacker News 热门（buzzing.cc 中文翻译）精选

AI 评分 77/100

开发者用 Claude Fable 5 在 Claude Code 中将 1993 年 Amiga 游戏 Babylonian Twins 移植到 Godot

作者让 Claude Fable 5 在 Claude Code 中分三步移植其 1993 年 Amiga 游戏：34,000 行 C++ 一个晚上迁入 Godot 4，72,758 行无注释 68000 汇编先用 vasm 重建出与发售版字节一致的二进制再移植，并把 1993 原作作为第二启动项嵌入新游戏。

智能体推理教程/实践编码

推荐理由：作者亲历者复盘用 LLM 移植 68000 汇编的完整过程，给出可验证的字节级校验方法和多处 AI 出错的实例。

02:29

OpenAI：官网动态（RSS · 排除企业/客户案例）精选

AI 评分 88/100

OpenAI 发布 GPT-6 Astra：多项基准刷新纪录， cybersecurity 能力达 Critical 阈值

OpenAI 发布新一代模型 GPT-6 Astra，称其在计算机使用、软件工程、科学和网络安全等方向达到 SOTA。

智能体 OpenAI 推理模型发布

另有 7 家信源报道

推荐理由：官方发布给出多项评测数字、定价和可用渠道，读者可以据此比较它相对前代和竞品的能力与成本变化。

00:00

Tessl：产品与工程博客精选

AI 评分 62/100

Tessl 提出反馈回路工程：用内、中、外三层循环让项目对智能体更友好

Tessl 提出反馈回路工程这一实践，把围绕智能体的循环分为内循环（测试、类型、评审）、中循环（维护智能体修复项目本身的状态）和外循环（读取生产信号），认为常被忽略的中循环能改善未来大量变更的条件。

智能体大佬观点编码

推荐理由：文章提出反馈回路工程的三层框架，并给出 Tessl 自身维护智能体的运行数据，可作为改进智能体开发环境的可迁移方法。

9月3日9月3日周四

星期四 · 4 条

19:29

Hugging Face：Blog（RSS）精选

AI 评分 72/100

Hugging Face 发布开源工具 funes，为编码智能体提供可本地持有的记忆层

Hugging Face 发布开源工具 funes，为 Claude Code、Codex、pi、Hermes 等编码智能体提供本地记忆层，把已有会话记录索引成 Lance 数据集，一条 funes add 命令即可让 Agent 自主召回原始出处（Agent、时间戳、会话、轮次）。

智能体 Hugging Face 产品更新开源生态

推荐理由：原文给出本地记忆层的检索机制、跨 Agent 复用方式，以及召回比压缩和交接更省成本的基准数字，方法可直接复用。

02:29

GitHub Blog精选

AI 评分 61/100

GitHub Copilot 如何在不牺牲任务质量的前提下降低 AI 编码成本

GitHub 工程师 Erik Kristensen 分享了 Copilot 降本的四项改动：选择性压缩工具输出、移除 view 工具行号前缀（线下推理成本降约 5%，线上用户日均推理成本降约 3%）、压缩 task-tool 提示词（每轮省约 1300 token，每活跃小时归一化成本降 2.9%）、后台任务完成后直接交付结果（AI Credits 用量降约 2.3%）。

智能体教程/实践编码部署/工程

推荐理由：GitHub 用四项改动说明压缩 token 为何要看整个任务而非单次调用，并给出可复用的评估方法与踩坑教训。

02:25

Cursor Blog精选

AI 评分 67/100

Cursor 推出 Self-Hosted Machines，云智能体可在企业自有机器上执行

Cursor 发布 Self-Hosted Machines，让云智能体的工具执行迁移到企业自有网络内的机器，智能体循环、推理和规划仍留在 Cursor 云端，通过 worker 的出站 HTTPS 连接对接，Cursor 不会主动连入企业网络。

智能体产品更新编码部署/工程

推荐理由：官方介绍了让智能体工具执行迁入企业自有基础设施的方案与适用场景， teams 可据此判断何时值得自托管以及如何按需扩容。

01:01

Claude：Blog（网页）精选

AI 评分 63/100

Anthropic 发布 Claude 商务智能体蓝图，含购物与商家智能体参考实现

Anthropic 发布面向电商、旅游、电信和票务平台的商务智能体蓝图，提供可在数天内上线的 harness、模式和护栏，附购物智能体与商家智能体的完整参考实现及 Claude Code 插件。

智能体产品更新搜索编码

推荐理由：原文给出完整可部署的购物与商家智能体实现、接入方式和合作生态，工程团队可据此评估落地铁路线。

AI 编码

精选归档 · 第 2 页

9月10日9月10日周四

9月9日9月9日周三

9月6日9月6日周日

9月5日9月5日周六

9月4日9月4日周五

9月3日9月3日周四

相关主题