Your AI chatbot isn't hallucinating. It's reading your outdated docs.

Description and use

Chatbots frequently provide incorrect answers due to outdated or contradictory documentation. The Knowledge Freshness Audit skill for Claude identifies and prioritizes these issues, enhancing answer accuracy.

Problem

Support chatbots often give incorrect answers because they rely on outdated or contradictory documentation rather than the model itself.

Outcome

The audit skill successfully identified all planted problems in a test set of four documents, demonstrating its effectiveness in detecting contradictions, outdated content, duplicates, unfinished documents, and gaps.

Workflow

• 1. Install the Knowledge Freshness Audit .skill file in Claude. 2. Attach the relevant SOPs, policies, FAQs, or help articles. 3. Prompt Claude with "Audit these help docs - our chatbot keeps giving wrong answers." 4. Review the generated report that lists contradictions, outdated files, duplicates, unfinished documents, and gaps. 5. Resolve the flagged issues in the source documents. 6. Re‑run the audit to confirm all problems are fixed.

Evidence

I built a Claude skill called Knowledge Freshness Audit. You attach your SOPs, policies, FAQs, or help articles, and it returns a prioritised fix list:

Source

I tested it on four small documents with planted problems: two contradictions (refund window and support email), three undated files, one duplicate, three files with no owner, one "TBD" answer, and one unanswered question. The skill found all of them.

Source

It never decides which fact is correct. If one page says 7 days and another says 14, the skill flags it and asks.

Source

It keeps personal data out of the report. It refers to files and lines instead of copying customer names, phone numbers, or account details.

Source