ChatGPT Import
PassAudited by ClawScan on May 1, 2026.
Overview
This skill appears to do what it says, but it is privacy-sensitive because it sends ChatGPT conversation text to OpenAI and stores the history in persistent OpenClaw memory.
Install/use this only if you intentionally want OpenClaw to search your past ChatGPT conversations. Before running it, review and remove secrets or sensitive chats, use a limited OpenAI API key, and remember that the generated SQLite database contains searchable conversation text.
Findings (3)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
Old chats, including secrets or outdated instructions, may be surfaced in future OpenClaw memory searches.
The main function is to place prior ChatGPT conversations into OpenClaw's searchable memory, which is purpose-aligned but creates persistent reusable context from potentially sensitive or instruction-like past chats.
Import your ChatGPT conversations into OpenClaw so they're searchable via memory search.
Filter sensitive conversations before import, keep the generated database only as long as needed, and treat retrieved chat history as reference material rather than authoritative instructions.
Conversation text may be processed by OpenAI's API, and embedding calls may incur API costs.
The skill clearly discloses that conversation content is sent to an external provider for embedding; this is expected for the stated workflow but privacy-sensitive.
The embed script sends conversation text to `api.openai.com` for embedding.
Only run this on data you are comfortable sending to OpenAI, remove secrets first, and consider a scoped or low-limit API key.
The API key can spend against the user's OpenAI account; passing keys on the command line can also expose them in shell history or process listings.
The script requires an OpenAI API key, either from an argument or environment variable, to call the embeddings API. This is disclosed and purpose-aligned, and the code does not show key storage.
api_key = args.api_key or os.environ.get('OPENAI_API_KEY')Prefer the OPENAI_API_KEY environment variable over --api-key, use a limited/scoped key if available, and avoid sharing logs or command history containing secrets.