AI Extension · Retrieval

Add Mini RAG

Start small with document-grounded answers. The goal is not a plausible answer, but an answer with evidence and citations.

Start small

Do not ingest every document first. Start with the documents needed to answer 20 frequent questions.

Mini RAG Implementation Prompt

Copy and use

"I want to add a mini RAG feature to an existing web service. Create an implementation plan and file-change outline from the project information below. [Project] - Frontend: [React/Vue/other] - Backend: [FastAPI/Hono/other] - Database: [Postgres/Supabase/other] - Document type: [Markdown/PDF/Notion/other] - Answer policy: [citations required/permission filter/answer unknown when missing evidence] [Output requirements] 1) document intake/chunking/embedding/storage flow 2) /ai/ask API contract 3) retrieval quality eval cases 4) UI pattern for citations 5) security/permission/cost checklist"

Copy this prompt and paste it into your AI chat.

API Contract Starting Point

api-contract.txt

POST /ai/ask
{
  "question": "How do I deploy this?",
  "filters": { "project": "choorai" }
}

Execution Checklist

0/5 done

Pick document sourceStart with a small source set

Choose chunking strategyStart from headings and paragraphs

Pick vector storageLocal, pgvector, or managed SaaS

Connect search APIReturn answer and sources together

Write retrieval evalsMeasure evidence recall, not just final answer

TL;DR

Use this section as a quick baseline for mini RAG.
Replace branch names, secrets, URLs, and thresholds with your team-specific values.

Prerequisites

Confirm repository and runtime access permissions.
Verify baseline commands (`install`, `build`, `test`) for local/CI.
Ensure environment variables, secrets, and deployment targets are up to date.

Steps

Lock the change goal and scope in one sentence.
Apply the core mini RAG changes in minimal commits.
Capture validation logs and immediately update checklist/docs.

Validation

Confirm build/test pass reproducibly on the current branch.
Run manual sample checks on critical user/operation flows.
Cross-check ko/en docs and links for intent consistency.

Troubleshooting

Capture logs, request payloads, and environment values first on failures.
Narrow down by permissions, env vars, routing, and build artifacts.
Record temporary mitigation and root-cause fix separately.

References

RAG Basics

Retrieval augmented generation flow

Vector DB

Embedding storage and search

Last updated: June 22, 2026 · Version: v0.0.1

Send Feedback

Opens a new issue page with your message.

Open GitHub Issue