What changed in Tech today
Articles, releases and code from Hacker News, Reddit, GitHub and the people building RPA, workflow automation and AI agents — plus what the community pushed to the top today.
-
01
The AI Agents Pitching Lit Mags to Stay Alive
AIAutonomous AI agents on iLands are cold-pitching literary magazines to earn tokens, illustrating how unconstrained outbound outreach workflows risk spamming prospects and burning brand reputation.
-
02
Built a deterministic IGA-style review packet for MCP/A2A agent capabilities
Redditr/AI_Agents 14m ago
-
03
if your no-code app has an AI chatbot in it, here's the test most people skip
Reddit -
04
Challenging anyone to actually get an LLM to write a complete Risch algorithm implementation
Reddit -
05
How do you test shorter agent instructions for changes in meaning?
Reddit -
06
mcp-safe-db: A strictly read-only SQL MCP server with engine-level write blocking
Reddit -
07
-
08
Is JEV (or similar) actually useful in a harness?
Reddit -
09
-
10
TryMCP Made It Possible to Connect Data to LLM in Minutes
Reddit -
11
Should a coding agent be allowed to change the tests it’s trying to pass?
Reddit -
12
Nous Research confirms it hit $1.5B valuation, launches AI agents for business users
AINous Research raised $90 million and launched Hermes for Businesses, enabling enterprises to deploy custom AI agents for multi-step workflows with private data controls.
ArticleTechCrunch Ent 1h ago
-
13
Claude Haiku 5.5
AIAnthropic released Claude Haiku 5.5 with default reasoning and lower base pricing, offering automation builders a cheaper fast model for tasks under 100,000 tokens.
ArticleSimon Willison 1h ago
-
14
Survey Finds AI-Generated Code Increases Debugging and Failure Rates and Creates a Comprehension Gap
AIA survey found AI coding agents increase debugging time and production failures, meaning automation builders must allocate more effort to testing and understanding agent-generated code.
ArticleInfoQ AI 1h ago
-
15
GPT-6 and Intelligent UI for everyone
ArticleOpenAI 1h ago