Build and launch websites and apps

Take a website or app from requirements to implementation, testing, security checks and deployment with a clear path to rollback.

Start with one working user journey, then add the screens and integrations it needs. Check responsive layouts, access controls and production behavior before expanding the release.

Markdown

• How to Build an Animated Website in Claude Code in an EveningRules against template-like design, visual directions, GSAP, accessibility, and performance testing for an animated website in Claude Code.
• Astro or Next.js: Which Should You Choose for an AI-Built Website in 2026A comparison of Astro and Next.js for a website built by an AI agent: purpose, structure, hosting, cost, and learning curve. Includes a scenario-based choice table.
• Is It Safe to Give Claude Code Access to Your Data in 2026In-depth analysis: Is it safe to give Claude Code access to your data in 2026?
• Vibe Coding Security: How to Protect an AI-Built AppHow to prevent an AI-built app from being hacked: the Lovable and Tea leaks, protecting secrets and keys, RLS, agent permissions, and a security checklist before publication.
• AI deleted the database: a 7-minute recovery planWhat to do after data deletion: stop the agent, document the incident, use git reflog, PITR, snapshots, and file recovery. Examples and safeguards.
• I built a product with AI—how to make improvements without breaking what worksSafe product maintenance after launch: git, rollback, staging, secret protection, hosting, backups, and upkeep. All six prompts are preserved.
• What Is Astro and Why Use It for an AI-Built Website in 2026Astro in simple terms: what this framework is, how it differs from Tilda, what kinds of sites it suits, and why you need it if AI is building your site.
• Vulnerability in Claude Code GitHub Action: What to Update and How to Protect Your RepoAnalysis of the claude-code-action vulnerability before v1.0.94: workflow checks, tool restrictions, least-privilege tokens, and CI/CD protection.
• Claude Design: A Guide and Five Practical ScenariosConnect Claude Code to Claude Design: export the design package, paste its ID into your project, and implement components. Includes features and limits.
• Codex Sites: Internal Tools and Service LimitationsHow Codex Sites, D1, and R2 work, who an internal site is for, how it differs from public site builders, and which limitations to consider.
• Coolify and Claude Code: Deploy a Project to Your Own VPSInstalling Coolify, connecting GitHub, deploying Next.js, domain and HTTPS, MCP, VPS requirements, and troubleshooting. Dollar-denominated technical prices and original limitations.
• Security Agents and Rollouts in Cursor: Security Checks and Release MonitoringHow to enable Security Reviewer, Vulnerability Scanner, and Rollouts in Cursor Automations: triggers, usage from the team pool, plans, and how they differ from Bugbot.
• Website Design with an AI Agent: Google Stitch and Claude CodeA complete manual Google Stitch to Claude Code workflow: export HTML/CSS, assemble DESIGN.md from local files, build and verify a page, and handle an unresolved MCP authentication step.
• How to Open Your Site to ChatGPT, Claude, and Perplexity Search CrawlersNames of search and training crawlers, a complete robots.txt example, sitemap, text availability, and citation limitations.
• How to check AI-generated code if you can’t read code: 7 checks for non-programmers in 2026AI wrote the code, but you don’t know how to read it? A ready-to-use checklist of 7 checks and prompts to catch broken code and security holes before publication.
• How someone else’s repository hacks Claude Code in 2026: 7 protection rulesFour documented indirect prompt-injection paths through repositories, GitHub content, Sentry events, and MCP metadata, with primary research and practical containment limits.
• Astro Site Through Claude Code: From Folder to PublicationStep-by-step guide: how to build a site with Astro through Claude Code without knowing how to program. What to install, which command to use to create a project, and how to publish it for free.
• Kimi Code Never Ask: modes and dangerous commandsIn Never Ask mode, Kimi Code runs rm -rf and shutdown without asking. How do the 3 modes differ, and how can you prohibit a dangerous command in config.toml?
• FrontierCode Diamond: 13.4% mergeable patches on 50 hard tasksThe June 2026 FrontierCode results: 13.4% on Diamond’s 50 hard tasks versus 51.8% on Extended, and how to inspect an AI-generated patch.
• Images and fonts in a product: sources and licensesUnsplash, Pexels, Google Fonts, and paid licenses: terms of use, attribution, and checking the rights for each file.
• GitHub Code in a Commercial Product: Licenses and What to CheckPermissive licenses MIT, Apache, and BSD, copyleft GPL/AGPL, the absence of a LICENSE file, and checking the origin of code added by AI.
• Website Design with an AI Agent: Claude Code and Refero DESIGN.mdRefero Styles design context: DESIGN.md structure, choosing a visual system, importing it into CLAUDE.md, four steps to create a page, and common mistakes.
• Security review in Claude Code: step-by-step guide and GitHub Action automation in 2026How to check the security of a vibe project with /security-review in Claude Code. Setup steps, a GitHub Action template, 6 types of vulnerabilities, and honest limitations.
• utm_source=chatgpt.com: how to measure traffic from ChatGPTUTM tag and referrer, a segment in Metrika, landing pages, GA4, and the roles of OAI-SearchBot, GPTBot, and ChatGPT-User.
• Vercel: Environment Variables Without .env in GitEnvironment Variables, Development/Preview/Production, a new deployment, and safely checking keys in Next.js without passing secrets to the browser.