--- title: "Blog — AI Code Security Insights | ShipSafe" description: "Security research, vulnerability analysis, and practical guides for developers responsible for production code they didn't fully write." doc_version: 2026-09-20 last_updated: 2026-09-20T19:42:20.346Z canonical: https://ship-safe.co/blog --- # Blog — AI Code Security Insights | ShipSafe ShipSafe Blog # Security research Vulnerability deep-dives, platform analysis, and practical guides for developers who own production code they didn't fully write. ![](https://ship-safe.co/_next/image?url=%2Fmascot%2Fbosun-scan.jpg&w=640&q=75&dpl=dpl_9WRuBGVFrmtjyKjqNw2VCigDq9zK) [LatestWindsurfMCPCVESecurityWindsurf's Zero-Click RCE: How an HTML Page Owns Your IDE (CVE-2026-30615)Windsurf processes attacker-controlled HTML, writes itself a malicious MCP server entry, and launches it — no clicks, no prompts, no warning. Here's exactly how CVE-2026-30615 works and what to do about it.May 22, 20267 min readRead article](https://ship-safe.co/blog/windsurf-zero-click-mcp-rce-cve-2026-30615) [ClineSupply ChainSecurityNPMClinejection: How One PR Compromised an AI Coding Agent Used by MillionsBetween Feb 17 3:26am PT and 11:30am PT 2026, every developer who installed cline@2.3.0 got a postinstall payload. Here's how it happened, how to check if you were hit, and the AI-tool supply-chain lessons.May 218 min read](https://ship-safe.co/blog/clinejection-supply-chain-attack)[Claude CodeGitHub ActionsPrompt InjectionSecurityComment and Control: One PR Title Drained Their Anthropic BillClaude Code, Gemini CLI, and Copilot Agent all run on pull requests. Researchers proved a single PR title can prompt-inject the agent and exfiltrate ANTHROPIC\_API\_KEY plus GITHUB\_TOKEN. Here's the attack and the seven-line fix.May 209 min read](https://ship-safe.co/blog/comment-and-control-pr-injection)[MCPSecurityAI AgentsMCP Tool Poisoning and Rug Pulls: The New Trust Problem in AI ToolsYou approve an MCP server on Monday. On Tuesday it changes its tool description. On Wednesday your agent silently exfiltrates emails. This is the rug pull (CVE-2025-54136). Here's the threat model and the defense.May 198 min read](https://ship-safe.co/blog/mcp-tool-poisoning-rug-pull)[CursorClaude CodePrompt InjectionSecurityThe Invisible Backdoor: Hidden Unicode in .cursorrules and CLAUDE.mdYour AI sees text humans can't. Zero-width and Unicode tag characters in agent config files let attackers embed prompt injection payloads in plain sight. Here's how to detect them and what to do.May 187 min read](https://ship-safe.co/blog/invisible-unicode-rules-file-attack)[CursorCVEGitSecurityCursor's Git Hook Trap: One Clone, Full RCE (CVE-2026-26268)An AI agent clones a repo. The agent runs git commit. A pre-commit hook the attacker placed fires with full developer privileges. No prompt, no click. This is CVE-2026-26268\. Here's the fix.May 176 min read](https://ship-safe.co/blog/cursor-git-hook-rce-cve-2026-26268)[LovableBreachSupabaseSecurityThe Lovable April 2026 Breach: What Got Leaked and What to CheckSource code, database credentials, AI chat history, and customer data — every pre-November 2025 Lovable project was exposed. Here's what we know, what was leaked, and the eight-step audit for your app.May 168 min read](https://ship-safe.co/blog/lovable-april-2026-breach)[LangChainCVESSRFSecurityThree LangChain CVEs in 30 Days: SSRF, Path Traversal, and SQL InjectionMarch 2026 disclosed three independent ways to drain secrets from LangChain. RecursiveUrlLoader redirects past SSRF guards. load\_prompt traverses paths. LangGraph SQLite checkpoints take SQL injection. Here's every CVE and patch.May 159 min read](https://ship-safe.co/blog/langchain-cves-2026-ssrf-path-traversal)[GitHub ActionsPwn RequestClaude CodeSecurityPwn Request Meets AI Agents: How GitHub Workflows Leak Your Secretspull\_request\_target plus actions/checkout of the PR ref equals secrets in the attacker's hands. Add a Claude Code agent on top and you get exfil from a single PR. Here's the pattern and the fix.May 148 min read](https://ship-safe.co/blog/pwn-request-ai-agents)[LLMSecurityCost ControlVibe CodingYour $4,200 Weekend: When Prompt Injection Drains Your Anthropic BillUber exhausted its 2026 AI budget months into the year. One developer burned $4,200 in three days on an autonomous refactor. Without max\_tokens and rate limits, your AI app is one prompt-injection away from the same. Here's the architecture.May 137 min read](https://ship-safe.co/blog/ai-cost-exhaustion-attack)[CursorSecurityCVEGuideCursor Security Risks: CVEs, Prompt Injection, and Code Vulnerabilities (2026)Three CVEs, prompt injection attacks, and auto-run exploits — plus what peer-reviewed research finds in deployed AI-built apps. Here's every Cursor security risk and how to protect yourself.Mar 3110 min read](https://ship-safe.co/blog/cursor-security-risks)[v0VercelNext.jsSecurityv0 by Vercel: 4 Security Gaps in Every Generated App (And the Fixes)v0 generates beautiful Next.js UI fast — but skips server-side validation, leaks API routes, and trusts client state. Here's what to check before you deploy.Mar 287 min read](https://ship-safe.co/blog/v0-vercel-security-risks)[ReplitSecurityGuideReplit Agent Security Guide: What It Misses and How to Fix ItReplit Agent builds and deploys full apps in minutes. But it consistently skips auth middleware, hardcodes secrets, and leaves debug endpoints live. Here's the complete fix guide.Mar 288 min read](https://ship-safe.co/blog/replit-agent-security-guide)[SupabaseRLSSecurityVibe CodingYour Supabase App Has No Row Level Security: A Vibe Coder's Fix GuideMost AI-generated Supabase apps ship with RLS disabled. That means anyone with your project URL can read, write, and delete your entire database. Here's how to lock it down in 15 minutes.Mar 286 min read](https://ship-safe.co/blog/supabase-rls-vibe-coding)[CursorSecurityResearchIs Cursor Safe? What the Research Actually ShowsPeer-reviewed research on 200 deployed AI-built apps found 90% carried at least one vulnerability, with broken access control in 75.5%. Here's what that means for Cursor-generated code, and how to fix the common classes.Mar 246 min read](https://ship-safe.co/blog/is-cursor-code-secure)[LovableSupabaseSecurity5 Security Vulnerabilities Every Lovable App Has (And How to Fix Them)Lovable builds beautiful Supabase apps fast. But it consistently misses Row Level Security, leaks service role keys, and more. Here's the fix for each.Mar 247 min read](https://ship-safe.co/blog/lovable-security-vulnerabilities)[Bolt.newSecurityGuideBolt.new Security Guide: How to Ship Without Getting HackedBolt.new generates full-stack apps in minutes. But without auth middleware and input validation, you're one exploit away from a breach. Here's the complete security guide.Mar 248 min read](https://ship-safe.co/blog/bolt-new-security-guide)[AI SecurityResearchVibe CodingAI-Generated Code Security: The Risks Nobody Talks AboutStanford research shows developers using AI assistants write significantly less secure code. Here's why, what types of bugs AI creates, and what you can do about it.Mar 249 min read](https://ship-safe.co/blog/ai-generated-code-security-risks)[ChecklistSecurityVibe CodingThe Vibe Coding Security Checklist (2026): Ship Fast, Stay SafeA complete security checklist for developers shipping AI-built apps. 20 checks across secrets, auth, injection, XSS, and configuration. Print it, pin it, use it.Mar 245 min read](https://ship-safe.co/blog/vibe-coding-security-checklist) ## Sitemap Every page of this site, in markdown: [https://ship-safe.co/sitemap.md](https://ship-safe.co/sitemap.md)