Tools & retrieval
What is Knowledge Base?
A knowledge base is the corpus an agent may retrieve from: docs, tickets, or policies.
The base must be curated and versioned. Stale or conflicting sources create confident wrong answers.
Why it matters
A knowledge base is the corpus an agent retrieves from. Its quality — freshness, accuracy, coverage, and chunking — determines the quality of every RAG answer.
A stale or poorly organized knowledge base produces confident wrong answers. Knowledge base maintenance is ongoing work, not a one-time setup.
Building a good knowledge base
Good knowledge bases have: versioned documents with timestamps, consistent formatting for reliable chunking, metadata tags for filtering, regular freshness audits, and a process for resolving conflicting information.
The most common failure is loading documents and forgetting them. Knowledge bases need the same maintenance discipline as production databases.
Key takeaways
- 1Knowledge base quality = RAG quality. Stale docs = wrong answers.
- 2Version, timestamp, and tag every document.
- 3Knowledge bases need ongoing maintenance, not one-time setup.
Common mistakes
- ✕Loading documents once and never updating them.
- ✕Not handling conflicting information between documents.
- ✕Chunking documents arbitrarily instead of by semantic boundaries.
Related terms
Concept neighborhood
Terms linked from Knowledge Base in the glossary graph.