Urgent.News

What's breaking now, across thousands of outlets.

AI

How to make Claude answer based on the osFoundry knowledge base and how to make it cite the article behind every claim

By the end of this guide, Claude (either the claude.ai chat with the connector, or Claude Code in your terminal) will be able to answer support and product questions from one of your osFoundry knowledge bases. Each claim it makes will point to an article ID and the time of the last update of that article. The next day Claude can ask osFoundry what has changed since then, and tell you which answers from yesterday were pinned on articles that have…

Translated from Hindi Read in Hindi

Claude, a chatbot available on claude.ai or through Claude Code in the terminal, will answer support and product questions using information from an osFoundry knowledge base. Its responses include article IDs and update times for each claim, allowing users to verify the information. The next day, Claude can check if any of the articles it cited have been updated, and if so, update its responses accordingly. Users can also edit articles and Claude can reflect those changes, but only with permission.

Written by urgent.news from Dev.to's report — not a translation of it. Machine-written — may contain errors; check the original before relying on it.

This story

This is one outlet's version. Read the fullest account.

Read the original at dev.to →

More in AI

Can ChatGPT Read Your Website? How I Found a Block My robots.txt Didn't Show

I wanted to know if ChatGPT ever cites my blog. So I asked it a few questions my posts answer, like "Can I host WordPress on Vercel?". My post on that ranks on the first page of Google.

  • ChatGPT unable to access website, receives 403 error
  • Major search engines (Googlebot, Bingbot) can access site
  • Server-wide rule blocks AI crawlers, not robots.txt

How do you evaluate the quality and accuracy of responses generated by an LLM?

Asked in a GenAI Engineer interview. What they are really testing: Whether you can turn a vague idea of a "good answer" into a repeatable evaluation pipeline that also covers retrieved context and…

  • Create a golden dataset with representative queries and reference answers
  • Assess faithfulness, answer relevance, and correctness against references
  • Use LLM as judge, calibrate against human-labeled samples for scalability

More from Monday 9 November →