skills.rest
buildoak
buildoakCOMMUNITY
Explore all skills by this author →
GitHub

chatgpt-search

Index ChatGPT conversation exports into SQLite FTS5 with BM25 ranking.

15|1|Updated Feb 17, 2026
Search Tags:#search#agent#sqlite#conversation#multilingual#tf-idf#fts5
One-click install
npx skills add https://github.com/buildoak/fieldwork-skills --skill chatgpt-search
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: chatgpt-search
Source: https://github.com/buildoak/fieldwork-skills/tree/main/skills/chatgpt-search
Command: npx skills add https://github.com/buildoak/fieldwork-skills --skill chatgpt-search

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires scikit-learn, langdetect, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This skill enables fast, offline search of ChatGPT conversation exports by indexing chats into a SQLite FTS5-powered search engine with BM25 ranking and TF-IDF keyword extraction.

Core Features & Use Cases

  • End-to-end search: index conversations, filter by date, role, model, language, and browse results.
  • TF-IDF enrichment: surface keywords per conversation to aid discovery.
  • Agent-ready: designed for Claude Code and Codex CLI to autonomously invoke the skill.

Quick Start

Run the setup script with your conversations.json to build and index the search database.

Frequently Asked Questions about chatgpt-search

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I search my ChatGPT conversation exports offline?▼

You can search ChatGPT conversation exports offline by indexing the conversations.json file into an SQLite FTS5 engine, which enables fast text queries with BM25 ranking without needing an internet connection.

Can I filter ChatGPT chats by date, model, or language?▼

Yes, the search engine supports filtering ChatGPT chats by date, role, model, and language, utilizing langdetect for multilingual identification to help you retrieve specific conversations accurately.

How does TF-IDF enrichment work for ChatGPT chat discovery?▼

TF-IDF enrichment analyzes your indexed ChatGPT chats to surface and extract the most important keywords per conversation, making it easier to discover and track relevant discussion topics across your history.

What is the best way to index a large ChatGPT export for full-text search?▼

The best way to index a ChatGPT export is running a setup script that parses the data, detects languages, and builds a SQLite FTS5 database with BM25 ranking and robust error handling for reliable full-text search.

Do I need scikit-learn and langdetect to search my ChatGPT history?▼

Yes, you need scikit-learn and langdetect installed, as they provide the TF-IDF keyword extraction and multilingual language detection capabilities required to enrich and categorize the search index.

Can Claude Code or Codex CLI autonomously search my ChatGPT chats?▼

Yes, this search pipeline is agent-ready and designed for Claude Code and Codex CLI to autonomously invoke, allowing the agents to parse exports, build the SQLite index, and retrieve past discussions.

Related Skills

Explore More Software Engineering Skills→
raydocsraydocs

flow-next-export-context

Export RepoPrompt context to a consolidated markdown file for external LLM review.

Community
Advanced
Arupbiswas09Arupbiswas09

distributed-search

Implement relevance-based text search with inverted index and TF-IDF/BM25 ranking.

Community
Advanced
IOchairIOchair
66

sql-manything

Indexes source code into SQLite and extracts snippets via SQL+FTS5 searches.

Community
Advanced
nadicodeainadicodeai

qmd

Index and search local text collections with hybrid retrieval.

Official
Advanced