zeabur-rag-learn

name

zeabur-rag-learn

description

Contribute new verified knowledge to the RAG knowledge base. Use after you've solved a concrete problem, debugged a non-obvious issue, or discovered undocumented behavior that would help future agents — but only after calling zeabur-rag-search first to confirm the knowledge isn't already covered. Only contribute verified facts, not guesses or workarounds. The new chunk is marked unverified until an admin reviews it.

RAG Learn

Add new knowledge to the knowledge base. The chunk is immediately searchable but marked as unverified until manually reviewed in the admin dashboard.

Base URL: $ZEABUR_RAG_URL Auth: Authorization: Bearer $RAG_API_KEY

API

curl -s -X POST "$ZEABUR_RAG_URL/api/learn" \
  -H "Authorization: Bearer $RAG_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "title": "Short descriptive title",
    "content": "The knowledge content in Markdown",
    "tags": ["relevant", "tags"],
    "source_query": "original user query"
  }'

Field	Required	Description
title	Yes	Short descriptive title
content	Yes	Knowledge content (Markdown supported)
tags	No	Array of tag strings for categorization
source_query	No	Original user query that prompted this knowledge

Returns { "id": "LEARNED-...", "ids": ["LEARNED-...-0", ...], "chunk_count": N, "status": "indexed", "verified": false }.

Long content is split into multiple chunks automatically; ids[0] is the first (parent) chunk, subsequent chunks link back via parent_id. id matches ids[0] for backward compatibility.

Guidelines

Only contribute knowledge that is:

• Verified correct — not a guess or workaround
• General enough — useful to future users, not just this one case
• Not already covered — search first to avoid duplicates