Skip to content
TakoAPI
Back to skills

chatgpt-import

by samdickson22

Coding Agents & IDEs
480Downloads reported by clawskills.sh
2Stars
0Views
0 likesSign in to like this skill
—Not yet rated
Your rating

Brief

Import ChatGPT conversation history into OpenClaw's memory search.

Description

What This Skill Does

Converts exported ChatGPT conversation history into a local SQLite database with vector embeddings, then registers it as an extra memory search path in OpenClaw. After setup, OpenClaw's memory search covers both its native memory and all past ChatGPT conversations.

Embeds conversations once into a local SQLite file, so all past ChatGPT history is queryable alongside native OpenClaw memory without re-uploading or reprocessing on each search.

When to Use It

  • Searching old ChatGPT conversations by topic or keyword
  • Migrating from ChatGPT without losing conversation history
  • Finding code snippets discussed in past ChatGPT sessions
  • Building a searchable personal archive of AI conversations
  • Recovering context from months-old ChatGPT threads

Example Workflow

INPUT

User asks: find my old ChatGPT conversations about database indexing

AGENT
  1. 1

    User exports conversations.json from ChatGPT account settings

  2. 2

    Runs convert_chatgpt.py to convert conversations into markdown files

  3. 3

    Runs bulk_embed.py with OPENAI_API_KEY to embed and store results in chatgpt-memory.sqlite

  4. 4

    Adds the SQLite path to OpenClaw config under memorySearch.extraPaths

  5. 5

    Restarts the OpenClaw gateway to load the new search path

OUTPUT

Memory search returns matching segments from past ChatGPT conversations alongside native OpenClaw results

Requirements

  • OPENAI_API_KEY environment variable
  • ChatGPT account with exported conversations.json

Installation

clawhub install samdickson22/chatgpt-import

Run the CLI command, or paste the prompt into Claude Code, Cursor, Codex or whichever coding agent you use.

Links