Urgent.News

600+ sources. One page. See who else covered it.

Editions

AI

iris-agentic-dev -- Give Your AI a Live Connection to IRIS, Part 1: The Problem, the Tool, and Getting Started

Part 1 of a series. Part 2 covers the full tool catalog. Part 3 covers ObjectScript skills. Part 4 covers benchmarking and measuring what actually improves. The Problem Hiding in the Comments Thomas Mazur's post "Frogs, Chickens, AI, and VS Code" on VS Code productivity — Peacock, scoped workspace files, Copilot Agent mode — drew a sharper problem in the comments. Pietro Di Leo and Mike.W pointed…

Thomas Mazur's post on VS Code productivity highlighted a problem when working server-side in VS Code with IRIS. Pietro Di Leo and Mike.W pointed out that Copilot can only see the files open in the editor, not the virtual filesystem. John Murray recommended a project called iris-agentic-dev to address this issue and the first article explaining the problem, the tool, and how to get started appeared.

The AI cannot see the full namespace when working with an isfs:// workspace because it only streams classes to the editor on demand via the Atelier API. Iris-agentic-dev provides a live connection to IRIS, allowing AI assistants direct access to the entire namespace, including the ability to search, compile, execute SQL queries, introspect class definitions, run unit tests, and debug.

Built with the community, iris-agentic-dev is an MCP server that connects to IRIS through the Atelier REST API, providing tools for AI assistants to interact with a live IRIS instance. The tool works with GitHub Copilot, Claude Code, Cursor, and OpenCode, and is open source under the intersystems-community GitHub organization.

Written by urgent.news from Dev.to's reporting — not their text. Machine-written — it may contain errors, so check the original before relying on it.

Read the original at dev.to →

More in AI

OpenAI Paused Astra for Cyber Risk. Your Agent's Sandbox Escape Is the Same Problem, Smaller Scale

OpenAI paused internal work on its upcoming model, Astra, after evaluations suggested it may have crossed into "Critical" cyber capability territory, including potential autonomous zero-day…

  • OpenAI paused Astra development due to cyber risk concerns.
  • Similar sandbox escape vulnerabilities exploited by Anthropic, Meta, Moonshot.
  • Sentinel proposed to detect unauthorized AI agent behavior and system access.

One AI Answer, Eight Brands: Designing a Benchmark Without Multiplying the Evidence

I recently ran a small China AI benchmark for eight luxury-jewelry brands. The most interesting result was not a platform ranking. It was a disagreement between two kinds of visibility.

  • Piaget visible in all four answers with China channels, absent from wedding jewelry recommendations
  • Benchmark collected 12 valid raw answers, analyzed separately from raw answers
  • Distinct metrics for different buyer decisions prevent misleading aggregate AI visibility scores