@anhhieu.hehe: cung cấp lượng oxy dồi dào cho cá khoẻ mạnh, ae mua máy sủi mini này về xài đi #emhíumêcá #cácảnh #thuysinh #thuỷsinh

Em Híu -phụ kiện Cá review
Em Híu -phụ kiện Cá review
Open In TikTok:
Region: VN
Sunday 30 August 2026 06:31:38 GMT
6677
17
1
1

Music

Download

Comments

kocoteen101
Chan đê mấy thg bột 🌽 :
Máy giống của em
2026-09-01 17:30:49
0
To see more videos from user @anhhieu.hehe, please go to the Tikwm homepage.

Other Videos

Most people think building an AI agent requires a team of developers. It doesn’t always. You can start with a simple structure and let Claude handle more of the repetitive work. This 6-step workflow shows how: 𝟭. 𝗜𝗻𝘀𝘁𝗮𝗹𝗹 𝗖𝗹𝗮𝘂𝗱𝗲 𝗖𝗼𝗱𝗲 The foundation everything else sits on. Pick whichever surface fits your workflow, set it up in under a minute, and everything below it compounds. 𝗧𝗲𝗿𝗺𝗶𝗻𝗮𝗹 macOS, Linux, WSL the original npm i -g 𝗗𝗲𝘀𝗸𝘁𝗼𝗽 𝗮𝗽𝗽 Native Mac & Windows window download 𝗩𝗦 𝗖𝗼𝗱𝗲 Side panel inside your editor extension 𝟮. 𝗕𝘂𝗶𝗹𝗱 𝗰𝗼𝗻𝘁𝗲𝘅𝘁 ~/CLAUDE.md is the onboarding doc Claude reads before every chat. Write it once, every session loads it. ## Role  ## Voice  ## Banned words  ## Defaults ▶ 𝗣𝗔𝗦𝗧𝗘 𝗧𝗛𝗜𝗦 > Help me build my CLAUDE.md from scratch. Use Boris Cherny’s CLAUDE.md as a starting template. Ask me about my business, voice, banned words, output defaults, and how I want you to work. Save the final file to ~/CLAUDE.md. 𝟯. 𝗕𝘂𝗶𝗹𝗱 𝗺𝗲𝗺𝗼𝗿𝘆 Every correction becomes its own .md file. The same mistake never lands twice. You correct → Claude saves .md → Loads next time ▶ 𝗣𝗔𝗦𝗧𝗘 𝗧𝗛𝗜𝗦 > From now on, whenever I correct you, save it as its own .md file at ~/.claude/projects/{project}/memory/, prefixed feedback_, user_, project_, or reference_. Index everything in MEMORY.md. 𝟰. 𝗕𝘂𝗶𝗹𝗱 𝘀𝗸𝗶𝗹𝗹𝘀 One command fires a whole workflow across multiple tools, no retyping a 200-word prompt every time. /your-skill → Notion + Gmail + Drive → Output ▶ 𝗣𝗔𝗦𝗧𝗘 𝗧𝗛𝗜𝗦 > Turn this workflow into a / {name} skill: [describe trigger, inputs, outputs]. Create ~/.claude/skills/{name}/SKILL.md, and wire the MCP connectors I need (Notion, Gmail, Drive, etc.), and make it fire-able from any chat. 𝟱. 𝗕𝘂𝗶𝗹𝗱 𝗮𝗴𝗲𝗻𝘁𝘀 Each agent gets one file and one job. Opus on the analysing roles, Sonnet on the executing roles, with a 95/100 QA gate that fails any work below the bar. STRATEGIST — Opus → BUILDER — Sonnet → QA GATE — 95/100 ▶ 𝗣𝗔𝗦𝗧𝗘 𝗧𝗛𝗜𝗦 > Build me an agent team for [process]. Create one .md file per specialist at ~/.claude/agents/. Use Opus for analysers, Sonnet for executors. Add a 95/100 QA gate that fails any work below the bar. 𝟲. 𝗥𝘂𝗻 𝗼𝗻 𝗮𝘂𝘁𝗼𝗽𝗶𝗹𝗼𝘁 Claude Routines turn your agent team into a cron job on Anthropic’s cloud, so you set it once and walk away. ◉ Daily 8am UK · runs in cloud ◉ Weekly Monday · drops to Notion ▶ 𝗣𝗔𝗦𝗧𝗘 𝗧𝗛𝗜𝗦 > Create a Claude Routine that runs my [agent or skill] every [schedule]. Run it on Anthropic’s cloud, not my laptop. Drop the output into [folder, Notion page, or tool].
Most people think building an AI agent requires a team of developers. It doesn’t always. You can start with a simple structure and let Claude handle more of the repetitive work. This 6-step workflow shows how: 𝟭. 𝗜𝗻𝘀𝘁𝗮𝗹𝗹 𝗖𝗹𝗮𝘂𝗱𝗲 𝗖𝗼𝗱𝗲 The foundation everything else sits on. Pick whichever surface fits your workflow, set it up in under a minute, and everything below it compounds. 𝗧𝗲𝗿𝗺𝗶𝗻𝗮𝗹 macOS, Linux, WSL the original npm i -g 𝗗𝗲𝘀𝗸𝘁𝗼𝗽 𝗮𝗽𝗽 Native Mac & Windows window download 𝗩𝗦 𝗖𝗼𝗱𝗲 Side panel inside your editor extension 𝟮. 𝗕𝘂𝗶𝗹𝗱 𝗰𝗼𝗻𝘁𝗲𝘅𝘁 ~/CLAUDE.md is the onboarding doc Claude reads before every chat. Write it once, every session loads it. ## Role ## Voice ## Banned words ## Defaults ▶ 𝗣𝗔𝗦𝗧𝗘 𝗧𝗛𝗜𝗦 > Help me build my CLAUDE.md from scratch. Use Boris Cherny’s CLAUDE.md as a starting template. Ask me about my business, voice, banned words, output defaults, and how I want you to work. Save the final file to ~/CLAUDE.md. 𝟯. 𝗕𝘂𝗶𝗹𝗱 𝗺𝗲𝗺𝗼𝗿𝘆 Every correction becomes its own .md file. The same mistake never lands twice. You correct → Claude saves .md → Loads next time ▶ 𝗣𝗔𝗦𝗧𝗘 𝗧𝗛𝗜𝗦 > From now on, whenever I correct you, save it as its own .md file at ~/.claude/projects/{project}/memory/, prefixed feedback_, user_, project_, or reference_. Index everything in MEMORY.md. 𝟰. 𝗕𝘂𝗶𝗹𝗱 𝘀𝗸𝗶𝗹𝗹𝘀 One command fires a whole workflow across multiple tools, no retyping a 200-word prompt every time. /your-skill → Notion + Gmail + Drive → Output ▶ 𝗣𝗔𝗦𝗧𝗘 𝗧𝗛𝗜𝗦 > Turn this workflow into a / {name} skill: [describe trigger, inputs, outputs]. Create ~/.claude/skills/{name}/SKILL.md, and wire the MCP connectors I need (Notion, Gmail, Drive, etc.), and make it fire-able from any chat. 𝟱. 𝗕𝘂𝗶𝗹𝗱 𝗮𝗴𝗲𝗻𝘁𝘀 Each agent gets one file and one job. Opus on the analysing roles, Sonnet on the executing roles, with a 95/100 QA gate that fails any work below the bar. STRATEGIST — Opus → BUILDER — Sonnet → QA GATE — 95/100 ▶ 𝗣𝗔𝗦𝗧𝗘 𝗧𝗛𝗜𝗦 > Build me an agent team for [process]. Create one .md file per specialist at ~/.claude/agents/. Use Opus for analysers, Sonnet for executors. Add a 95/100 QA gate that fails any work below the bar. 𝟲. 𝗥𝘂𝗻 𝗼𝗻 𝗮𝘂𝘁𝗼𝗽𝗶𝗹𝗼𝘁 Claude Routines turn your agent team into a cron job on Anthropic’s cloud, so you set it once and walk away. ◉ Daily 8am UK · runs in cloud ◉ Weekly Monday · drops to Notion ▶ 𝗣𝗔𝗦𝗧𝗘 𝗧𝗛𝗜𝗦 > Create a Claude Routine that runs my [agent or skill] every [schedule]. Run it on Anthropic’s cloud, not my laptop. Drop the output into [folder, Notion page, or tool].

About