Security guidance for AI-built apps.

Read LyraShield AI research and practical guidance on securing AI-built apps, interpreting security evidence, verifying findings, and retesting fixes.

Follow via RSS

Browse by category

6 categories

Builder workflowsVibe-Coding SecurityPractical security habits for builders shipping apps written largely by AI coding tools.92articles
Identity & data boundariesAccess ControlAuthentication, authorization, row-level security, and sessions — keeping the wrong people away from the right data.44articles
Browser attack surfaceWeb SecurityHeaders, CORS, injection, and the rest of the browser-facing attack surface of a modern web app.40articles
Dependencies & pipelinesSupply ChainDependencies, packages, CI/CD, and third-party services — securing the code you didn't write yourself.17articles
Autonomy & guardrailsAgent SecurityPermissions, prompt injection, sandboxing, and guardrails for coding agents and agent-native applications.24articles
Evidence & retestingVerificationChecking that findings are real and fixes actually hold: evidence, retesting, and honest security claims.91articles

More practical guidance

100 articles

An identity signal passing through a gate to one scoped resource

Jul 17, 20266 min read

Why Client-Side Auth Is Not Access Control

Understand why route guards and hidden controls cannot enforce access, then place authorization at the server operation that needs it.

Nested service modules contained inside a hardened platform boundary

Jul 17, 20267 min read

Secure Cloudflare Workers Built With AI

Treat Worker bindings as capabilities, keep secrets encrypted, constrain outbound fetch and caching, and verify routes with production-shaped Wrangler config.

A processing core inside an isolation chamber with one filtered egress path

Jul 17, 20266 min read

Secure an OpenAI Codex Project

Set Codex sandbox, approval, network, environment, and MCP boundaries, then verify code changes before any release action.

Jul 17, 20266 min read

Sandbox Coding Agents and Control Egress

Isolate coding agents with disposable workspaces, minimal mounts, non-root execution, resource limits, and explicit outbound network policy.

Data signals entering a constrained execution funnel while other paths are blocked

Jul 17, 20266 min read

Command Injection in Agentic App Workflows

Remove unsafe shell boundaries from agent tools, constrain process arguments, and isolate the external commands that remain necessary.

Client requests crossing origin, protected transport, and capacity gates

Jul 17, 20267 min read

CORS for Vibe-Coded Apps

Replace wildcard or reflected origins with a minimal browser-sharing policy, correct credential handling, and cache-safe headers.

An incident signal contained before entering a retest path

Jul 17, 20267 min read

What to do with a critical security finding

Treat critical severity as a triage input: confirm scope and exposure, contain credible risk, preserve evidence, fix, and run a fresh retest.

Jul 17, 20266 min read

CSRF Protection for Cookie-Based AI Apps

Protect cookie-authenticated state changes with framework controls or verified tokens, plus SameSite and Fetch Metadata layers.

A workflow crossing staged security gates and an exact approval lock before production

Jul 17, 20268 min read

Secure a Cursor-Built App Before Launch

Review Cursor data handling, project rules, dependencies, app controls, agent permissions, and independent evidence before launch.

An untrusted file or diagnostic channel isolated from executable and public-response space

Jul 17, 20267 min read

Remove Debug Routes and Verbose Errors

Stop production errors from exposing stack traces, paths, queries, and framework internals while preserving useful, access-controlled diagnostics.