Agent context needs a development lifecycle
Read original source →thenewstack.io
Skills, agent configurations, prompt instructions, and rules files now determine what coding agents produce, shaping every line of generated code. The article argues these artifacts are functionally software and need a proper development lifecycle.
1 source
More stories today
ChatGPT's Chrome extension can now read your open tabs
With the extension installed, ChatGPT can view open browser tabs when you tag Chrome in its web or desktop app, and ChatGPT Work can control tabs to send messages, fill forms, or download files. OpenAI's ChatGPT Atlas browser is gone.
The Verge·53 minutes ago
Meta Omnilingual ASR named a TIME Best Invention of 2026
AI at Meta·1 hour ago
Claude Code Review exits research preview, now generally available
Code Review is out of research preview for Claude Team and Enterprise plans, with most reviews now finishing about 3x faster (~20 minutes down to 6-10 minutes). Admins pick Quick, Standard or Deep per repository; eligible Enterprise orgs can claim at least $10,000 in credits by January 15, 2027.
Claude Blog·1 hour ago
Perplexity API now available inside Lovable apps
Aravind Srinivas·1 hour agoStudy finds ChatGPT, Claude, Gemini hide their own errors
Nav Toor·1 hour ago
Tinker post-training API marks one year since launch
Thinking Machines·1 hour ago
Grok Bot users report autonomous microbusinesses earning $1,800
Alex Finn says a Grok Bot built and maintained a microbusiness that made $1,800, after earlier posts citing $1,100, 56,000 visitors, 123,000 page views and 2 paying sponsors. A solo developer describes running support, engineering and security bots on a Linux server.
Alex Finn·1 hour ago
Databricks AI research team details its use of AI Runtime
Databricks published a first-party account of how its own AI research team uses AI Runtime to develop and evaluate new research techniques.
Databricks Blog·1 hour ago
