Vishakha Singhal

🚨Breaking: Anthropic engineers revealed a simple trick they use internally. Claude agents can now remember how to improve themselves. A file called: AGENT_LEARNINGS.md The AI updates this file whenever it makes a mistake. Inside it: • mistakes it made • patterns to avoid • better approaches Before starting new tasks, the agent reads the file. Result? The agent gets smarter over time without retraining the model. This is called external memory scaffolding. Expect this pattern to show up in every serious AI agent system. ♻️ Repost to share with your audience. ✔️ You can follow @swadeshkumar_ for more internal updates.

Most people think using Claude Code is about writing better prompts. It’s not. The real unlock is structuring your repository so Claude can think like an engineer. If your repo is messy, Claude behaves like a chatbot. If your repo is structured, Claude behaves like a developer living inside your codebase. Your project only needs 4 things: • the why → what the system does • the map → where things live • the rules → what’s allowed / forbidden • the workflows → how work gets done I call this: The Anatomy of a Claude Code Project 👇 ━━━━━━━━━━━━━━━ 1️⃣ CLAUDE.md = Repo Memory (Keep it Short) This file is the north star for Claude. Not a massive document. Just three things: • Purpose → why the system exists • Repo map → how the project is structured • Rules + commands → how Claude should operate If CLAUDE.md becomes too long, the model starts missing critical signals. Clarity beats size. ━━━━━━━━━━━━━━━ 2️⃣ .claude/skills/ = Reusable Expert Modes Stop repeating instructions in prompts. Turn common workflows into reusable skills. Examples: • code review checklist • refactoring playbook • debugging workflow • release procedures Now Claude can switch into specialized modes instantly. Result: More consistent outputs across sessions and teammates. ━━━━━━━━━━━━━━━ 3️⃣ .claude/hooks/ = Guardrails Models forget. Hooks don’t. Use hooks for things that must always happen automatically. Examples: • run formatters after edits • trigger tests after core changes • block sensitive directories (auth, billing, migrations) Hooks turn AI workflows into reliable engineering systems. ━━━━━━━━━━━━━━━ 4️⃣ docs/ = Progressive Context Don’t overload prompts with information. Instead, let Claude navigate your documentation. Examples: • architecture overview • ADRs (engineering decisions) • operational runbooks Claude doesn’t need everything in memory. It just needs to know where truth lives. ━━━━━━━━━━━━━━━ 5️⃣ Local CLAUDE.md for Critical Modules Some areas of your system have hidden complexity. Add local context files there. Example: src/auth/CLAUDE.md src/persistence/CLAUDE.md infra/CLAUDE.md Now Claude understands the danger zones exactly when it works in them. This dramatically reduces mistakes. ━━━━━━━━━━━━━━━ Here’s the shift most people miss: Prompting is temporary. Structure is permanent. Once your repository is designed for AI: Claude stops acting like a chatbot... …and starts behaving like a project-native engineer. 🚀 Follow @swadeshkumar_ for more !!

🚨Breaking: Anthropic engineers revealed a simple trick they use internally. Claude agents can now remember how to improve themselves. A file called: AGENT_LEARNINGS.md The AI updates this file whenever it makes a mistake. Inside it: • mistakes it made • patterns to avoid • better approaches Before starting new tasks, the agent reads the file. Result? The agent gets smarter over time without retraining the model. This is called external memory scaffolding. Expect this pattern to show up in every serious AI agent system. ♻️ Repost to share with your audience. ✔️ You can follow @swadeshkumar_ for more internal updates.

Patch notes I'd like to see: •Buff Mota bhai •Nerf Elon •Add more unhinged special attacks •Fix absolutely nothing This game is already perfect 😭 billionaire-arena.emergent.h… @elonmusk

If you design with AI, this is all you need 1. modulify.ai — cheaper than lovable 2. landdding.com — top notch landings 3. motionin.design — 1 prompts 1 website 4. wonderlist.design — 127 full figma designs 5. iconly.design — world best icons 6. fontinlogo.com — fonts behind brands 7. on.design — design community 8. godly.website — award-winning websites 9. 60fps.design — mobile UI animations 10. footer.design — footer inspiration 11. recent.design — next level design 12. backgrounds.supply — backgrounds save this for later. Follow for more

The API already told you what's wrong. Most developers just don't know how to read it. That's why they spend: ❌ 2 hours checking the backend ❌ 1 hour reviewing logs ❌ 30 minutes asking AI When the answer was already in the response. Here's the simplest API debugging framework you'll ever learn: 🟢 2xx = Everything worked 200 → Request succeeded 201 → Resource created 204 → Success, nothing to return Translation: "Your API is doing exactly what you asked." Stop debugging. 🔵 3xx = Look at the route 301 → Moved permanently 302 → Temporary redirect 304 → Cached version available Translation: "You're probably looking in the wrong place." - Check URLs. - Check redirects. - Check caching. 🟠 4xx = The request is wrong 400 → Bad request 401 → Missing authentication 403 → Permission denied 404 → Resource not found 409 → State conflict 422 → Validation failed 429 → Rate limited Translation: "The server understood you. It just doesn't like what you sent." Check: ✓ Headers ✓ Authentication ✓ Parameters ✓ Request body ✓ Rate limits 🔴 5xx = The server is broken 500 → Internal server error 502 → Bad gateway 503 → Service unavailable 504 → Gateway timeout Translation: "The problem is not on your side." Check: ✓ Server logs ✓ Infrastructure ✓ Third-party services ✓ Database connections ✓ Network issues The fastest debugging rule I know: 2xx → Celebrate 3xx → Follow the redirect 4xx → Fix your request 5xx → Check the server That's it. No fancy framework. No 500-page documentation. Just four categories. And suddenly API debugging becomes predictable. Resources to learn : → w3schools : Learn the fundamentals: • HTTP methods • Request/response cycle • Status codes • REST basics → MDN Web Docs : Understand how the web really works: • HTTP headers • Authentication • Caching • Error handling → Postman Learning Center : Learn by doing: • API testing • Collections • Environment variables • Automated requests Follow @swadeshkumar_ for more such content 📌 Save this before your next production incident. You'll need it sooner than you think. ♻️ Repost if you know a developer who still treats every error the same way. 🚀

Most people think using Claude Code is about writing better prompts. It’s not. The real unlock is structuring your repository so Claude can think like an engineer. If your repo is messy, Claude behaves like a chatbot. If your repo is structured, Claude behaves like a developer living inside your codebase. Your project only needs 4 things: • the why → what the system does • the map → where things live • the rules → what’s allowed / forbidden • the workflows → how work gets done I call this: The Anatomy of a Claude Code Project 👇 ━━━━━━━━━━━━━━━ 1️⃣ CLAUDE.md = Repo Memory (Keep it Short) This file is the north star for Claude. Not a massive document. Just three things: • Purpose → why the system exists • Repo map → how the project is structured • Rules + commands → how Claude should operate If CLAUDE.md becomes too long, the model starts missing critical signals. Clarity beats size. ━━━━━━━━━━━━━━━ 2️⃣ .claude/skills/ = Reusable Expert Modes Stop repeating instructions in prompts. Turn common workflows into reusable skills. Examples: • code review checklist • refactoring playbook • debugging workflow • release procedures Now Claude can switch into specialized modes instantly. Result: More consistent outputs across sessions and teammates. ━━━━━━━━━━━━━━━ 3️⃣ .claude/hooks/ = Guardrails Models forget. Hooks don’t. Use hooks for things that must always happen automatically. Examples: • run formatters after edits • trigger tests after core changes • block sensitive directories (auth, billing, migrations) Hooks turn AI workflows into reliable engineering systems. ━━━━━━━━━━━━━━━ 4️⃣ docs/ = Progressive Context Don’t overload prompts with information. Instead, let Claude navigate your documentation. Examples: • architecture overview • ADRs (engineering decisions) • operational runbooks Claude doesn’t need everything in memory. It just needs to know where truth lives. ━━━━━━━━━━━━━━━ 5️⃣ Local CLAUDE.md for Critical Modules Some areas of your system have hidden complexity. Add local context files there. Example: src/auth/CLAUDE.md src/persistence/CLAUDE.md infra/CLAUDE.md Now Claude understands the danger zones exactly when it works in them. This dramatically reduces mistakes. ━━━━━━━━━━━━━━━ Here’s the shift most people miss: Prompting is temporary. Structure is permanent. Once your repository is designed for AI: Claude stops acting like a chatbot... …and starts behaving like a project-native engineer. 🚀 Follow @swadeshkumar_ for more !!