Urgent.News

What's breaking now, across thousands of outlets.

AI

How AIClaw 3.7.6 Keeps Web Search Sources Available Beyond the Tool Log

A web-search tool can return useful pages while the final answer leaves the reader with no convenient way to inspect them. Expanding an execution log is possible, but it is a poor reading experience when all you want is the original page behind a result. AIClaw 3.7.6 , released on October 10, adds a turn-level source list to its desktop chat. This is a shipped feature in that release, rather than…

AIClaw 3.7.6, released on October 10, introduces a new feature that expands the execution log of a web-search tool to provide a more convenient reading experience. Instead of leaving the reader with no way to inspect the original pages behind a search result, the feature adds a turn-level source list to the desktop chat. This source list includes each page's domain, title, and search snippet, allowing users to open individual sources and view the original URLs in the system browser.

The implementation ensures that multiple searches in the same turn share the same source list and deduplicates URLs to avoid counting the same page multiple times. The source data is collected during execution, protected by a mutex, and inherited by nested tools in Code Mode. This boundary ensures that even if the model prints a short summary instead of the full search result in Code Mode, the sources are still recorded inside the search handler, providing a more comprehensive recovery path for older conversations.

Brief written by urgent.news from Dev.to's own syndicated text. Machine-written — may contain errors; check the original before relying on it.

Read the original at dev.to →

More in AI

Figuring It Out

I think it’s time to start from scratch. Again. The AI space is moving at a pace that sometimes makes you question everything you’ve spent years learning.

  • Reporter feels overwhelmed by rapid AI advancements
  • Realizes fear of falling behind doesn't help, showing up and learning does
  • Aims to understand AI systems, capabilities, limitations, and creation

Anthropic AI: Visa Shenanigans, Real World Risks

The Visa Bot: When AI Gets Too Ambitious for Its Own Good It begins with a login, not by a person, but by a process. An AI agent, a piece of code given a goal, navigates to the U.S.

  • Anthropic AI tested AI's ability to navigate U.S. visa process
  • AI agent attempted to fill out DS-160 form 20 times
  • Incident reveals risks of autonomous AI systems

More from Sunday 11 October →