Ideas across the ecosystem
Three layers. One place to think, build, and share what we learn.
Choose a topic to explore AI agents, application architecture, runtime and rendering, or real product workflows. Browse twelve articles at a time, newest first.
Showing 1–12 of 12 articles
Understand agent context, inspect project structure, and keep your judgment when reviewing AI proposals.
MCP replaced its HTTP+SSE transport with Streamable HTTP. What changed, when to use stdio, Streamable HTTP or plain SSE, and how Semitexa uses stdio for MCP and SSE for the browser.
Choose the transport →Prompt management for PHP: prompt templates with ids, fail-closed rendering, a preview of what the model receives, per-tenant overrides and drift checks.
Open the prompt catalog →Broken access control and IDOR top the OWASP list, and AI agents forget the check. See how Semitexa makes every PHP route declare who may call it.
See the access contract →What design tokens are, how one token contract gives dark mode with light-dark(), and how to let an LLM restyle an app by choosing, never writing CSS.
Read the token contract →Harness engineering for PHP: guides and sensors around a coding agent, the PHP tools for each, running the right checks per change, and a Laravel starter.
Build the harness →Claude Code reads AGENTS.md since 2.1.277, but only without a CLAUDE.md. How to keep one instruction file for every agent, and how ai:verify keeps it short.
Keep one file →Can an external AI agent find reliable facts and a supported action on your site? Follow real Semitexa discovery documents step by step.
Follow an external agent into the site →PixelLeak shows how proof of a UI fix can expose private data. Follow a review through capture, storage, sharing and a publication boundary.
Follow the review artifact boundary →Cognitive surrender starts when a convincing AI patch replaces your judgment. Check it with Semitexa rules, Project Graph, traces and verification.
Review the evidence behind the proposal →See your project through the evidence an AI agent can retrieve. Walk a real route in Semitexa's visual Project Graph and learn what the map cannot prove.
See the project through an agent’s context →AI-native PHP development with Semitexa: trace a shipping bug through Project Graph, Observatory, and tests. Run the example and verify the result.
Read the AI-native PHP guide →Project Graph in Semitexa maps PHP routes, handlers, dependencies and impact. See real CLI queries, AI context, and working Demo workflows.
Read the Project Graph guide →What would help you move forward?
Your brief goes to our team. It isn’t published on the website. A high-level overview is enough to start.
Explore the offers and our work