与「Ask_gugudan」相关的搜索结果

[📣NOTICE] 11/19(월) 오후 4시🕓 Twitter Blueroom LIVE Q&A 진행 안내✔ 🖇2018.11.19 MON 4PM (KST) ▶구구단에게 궁금한 점들을 #구구단# #Ask_gugudan# 해시태그와 함께 보내주세요! (@gu9udan) #gugudan# #TwitterBlueroom#

Claude Code 推出实验功能 Code Mode：从一次调一个工具，到写一段 JS 批量调工具 Code Mode 默认关闭，要自己打开。打开以后，Claude 调工具的方式变了：以前每次只调一个 Bash、Read 或 Edit，看完结果再决定下一步；现在 Claude 先写一段 JavaScript，在里面一次调用多个工具，由 Claude Code 在本地会话里执行。 # 具体怎么工作 官方举例是把重试上限从 3 改成 5。Claude 会写一段代码：先用 rg 找到文件，再改参数，接着跑相关测试，最后只把文件路径和测试输出交还给自己看。原本要来回四五轮的操作，现在一步就能做完。 这样做有三个好处。几个独立的命令可以并行跑；上一步的结果可以直接传给下一步，不用经过模型；中间产生的大量输出可以先在代码里筛一遍，只把需要的部分放进上下文。官方推荐的用法也围绕这几点：同一处改动要改很多文件、搜索结果需要先过滤再读、几个独立的命令可以同时跑。 对话界面里看到的仍然是每一次工具调用，比如编辑的 diff。想看背后的代码，按 Ctrl+O 打开记录，展开 Code 行就能看到。 # 门槛和限制 开启方法是设置 CLAUDE_CODE_EXPERIMENTAL_CODE_MODE=1，可以在 shell 里临时设，也可以写进 ~/.claude/settings.json。写在项目的设置文件里不生效，所以克隆下来的仓库没法替你偷偷打开它。第一次开启时，Claude Code 会先请你确认。 限制不少： · 只能在终端的交互式会话里用，不支持 -p、Agent SDK、VS Code 面板、桌面端和云端会话。 · 不能走网关、Bedrock、Google Cloud 或 Microsoft Foundry，也不能用 API key helper。 · 只支持 Opus 5.5、Sonnet 5.5、Haiku 5.5 和 Fable 5.2。 · 开着 agent teams、OpenTelemetry 导出或 DISABLE_TELEMETRY 时无法启动。 · 用 Code Mode 开的会话，恢复时也必须开着 Code Mode。 · 代码在前台跑，最长 10 分钟；放到后台最长 2 小时，但要求会话在 auto 或 bypassPermissions 模式下启动。 # 权限怎么管 代码里的每一次工具调用，仍然要经过你的 deny、ask 规则和 PreToolUse hook。真正变化的是 auto 模式：分类器现在审的是整段代码，不再一次次审单个调用，而且看不到运行时才算出来的值，比如从搜索结果里取出的路径。官方明确建议，必须管住的调用要靠规则或 hook 兜底。想审代码本身，可以写一个匹配 REPL 的 hook，从 tool_input.code 读出 JS。企业可以在托管设置里把这个变量设成 0，全员禁用。

FAN SZN with @IPXExchange As a creator, do I ever hold the raised money directly? Only the part that's unlocked. The rest stays in escrow until its stage is approved. Picture a concert tour: Ticket sales sit in a box-office account the tour manager can't touch Each stage's installment opens only after that stage is signed off The tour manager takes one installment at a time, never the whole box at once Here's how it works, per IPX: 1. Raised capital never sits in a creator's wallet. It's held in a non-custodial smart contract escrow 2. Before the raise, the creator submits a line-by-line budget. IPX says its AI agent, Alfred, sets it up in a contract tied to milestones, and that CertiK audited it 3. After the project graduates, its budget moves into a separate escrow. A stage unlocks once the creator submits proof and the platform approves it, and then it's available for the creator to transfer My read: as a creator, I'd hold an installment only after I prove its stage, and never the whole raise up front. Real count, from IPX's text: it names 3 example stages: pre-production, filming and post-production. Made-up scenario, not a real project: a film uses those 3 stages. Pre-production is approved, so 1 of 3 installments is available to transfer. Filming and post-production stay in escrow until they pass their checks. My take: the first thing I'd ask is how long an approval takes. If a stage sits locked while I wait, that could hold up the whole shoot. What comes out the other end, per IPX? • Each stage's installment, after its milestone is approved • A record of every request, approval and release on IPX and Base, so supporters can follow along • What a creator pays, if anything: I didn't find it listed, so I'd check the project terms The honest part: • IPX says unlocks happen automatically with no human gatekeepers, and also says the platform approves each milestone. I couldn't find how those fit together or how long an approval takes • The Defiant labels the announcement "Presented by IPX," and I haven't seen anyone outside IPX confirm how it runs • I haven't seen it run on a live project yet So before I launch, I'd find out who approves each milestone and how long an approval takes. I'll want both answers before I confirm. Why this matters for creators and fans: → Creators know the payout path before they launch → Fans can see which stages are open and follow each release → Both sides work from the same milestones Source: IPX's router announcement and its KPI-Gated Escrow announcement Which would you check first: the budget lines or who approves each milestone?

My Watchlist got smaller this week, and that felt like progress. @GeckoTerminal shows me the list I actually use, not the list I wish I had, and keeping it short is the whole point for me. Every pool on there has a reason I can say out loud: what I am trying to learn from it, or what question it helps me answer. If I cannot say that reason clearly, it comes off. If I can, it stays, even when nothing is happening with it, because I am tracking to learn, not to chase movement. I check it the same way I tidy a desk. I look at what is still earning its place, I ask what has changed since I added it, and I let go of those that no longer teach me anything. That keeps my research honest, and it keeps my attention on the pools I already understand best. That is only how I keep mine, not advice for anyone else. My setup is here if you want a look: [your Watchlist link] What did you last take off your Watchlist, and why did it lose its place?

Who decides the answer to a prediction market? Usually the creator. Or an AI on a server nobody can check. So we built Noa Court - an AI resolver for Noa, live on NEAR mainnet. In the video we ask Gemini a vague question: "will the NEAR hackathon get more than 2000 projects?" - It finds the real source (Nearhackathon API) and samples it live: 1,803 right now - Turns the vague question into precise, hashed rules - We review the market and sign with a NEAR wallet - Market #27# goes live on Noa, published with readable rules How it resolves: - One Court contract, one pinned resolver running in an @out_layer TEE - Anyone can trigger it. Nobody can pick the answer - Numbers? Decided by code - Words? 3 NEAR AI judges must all agree and quote the evidence Asked "did a crypto hack make the news?", all 3 judges quoted a headline that never says "hack". It refuses to guess: - Prompt injection - flagged - Dead or changed source - void + refund Works with prices, NEAR tokens (powered by Intear & JustHoot), news feeds, web pages, any JSON API. Plug it into any market, contract or agent. Add the MCP to Claude, ChatGPT, Gemini - any AI that speaks MCP: Code: What would you never let an AI resolve? Built at NEAR HACKATHON @NEARProtocol @NearBuilders #NEARHackathon#

holy f*ck... Elon Musk's team just made airport rebooking lines useless my flight got cancelled and 187 people were ahead of me at the desk. my Grok Bot rebooked me before the line moved 17 people Grok Bot makes phone calls now. it sits on hold, talks to the agent and says up front it's an AI what it did while I stood in line: → sat through 11 minutes of hold music → got me on the 7:40 AM flight tomorrow → the airline cancelled, so it asked for a hotel and meals. got both → "aisle seat is $49 extra" it didn't pay. it asked me. I took the free middle seat → texted my wife the new landing time (signed Grok Bot) and set my alarm 14 minutes on the phone. I walked out of the line, 168 people still in it all it took was 3 rules I sent it once: 1. on every call, say you're an AI calling for me 2. ask for everything I'm owed, nothing made up 3. anything that costs money: ask me first the Grok Bot Research Desk setup is in my article. read it below ↓

🇺🇸🇨🇳 Rubio reportedly set Jensen Huang's downfall in motion... with a memo that hit harder than the CIA report it was based on The Washington Post reports the Secretary of State took a CIA report on Jensen's relationship with a suspected Chinese spy and rewrote it in his own, more forceful terms, according to one person familiar with it. When that memo's findings leaked, the scandal began. A former official who worked on chip policy described what Trump's national security team was after: "They wanted to destroy the influence Jensen had over the president." Trump still calls him a "legend." The agencies still can't prove she was a spy. But three bills restricting chip sales to China are now headed for the defense bill, and the one executive who used to talk Congress out of them is gone. So ask yourself who came out ahead this week, and you have a good idea why that memo found its way to reporters. Source: Washington Post / Writer: Daniel