Official scripts
Explore our curated collection of scripts
Who am I (ChatGPT)
Report which ChatGPT account the browser is signed into: id, name, email, picture and identity provider, or loggedIn:false when the browser is anonymous. Never returns the session's access token.
ChatGPT whoami (signed-in account)
Report which ChatGPT account this browser is signed in as: the user id, email and name. When the browser is signed out of ChatGPT, the run fails with a message asking you to sign in to chatgpt.com, so it can be used to check a session before running other ChatGPT scripts.
Get ChatGPT mode (Chat or Work)
Report whether ChatGPT answers new chats in Chat or Work mode in this browser, as the Chat/Work switch at the top of the home page shows it. The choice belongs to the browser, not the account, and a Chat choice lasts about four hours before ChatGPT falls back to Work, so check it before asking questions that need Chat. Requires a signed-in session; a signed-out browser fails with an error that says so. Use set_mode to change it.
ChatGPT switch display language
Switch the ChatGPT web app's display language via Settings, for a signed-in account. Requires an active session.
List ChatGPT models
List the models ChatGPT offers the signed-in account and the lanes its model picker is built from, plus the account's default model. Each model carries its slug, title and description; each lane carries its id, name, default model, the models it holds and the plan it requires. On a free account the picker itself is not shown and only the free lanes come back, which is what the site offers there. A signed-out browser fails with a signed-out error asking to sign in to chatgpt.com.
Get ChatGPT Codex usage limits
Read the Codex usage limits of the ChatGPT account open in this browser, as ChatGPT loads them for its Codex usage panel: the plan, whether Codex use is allowed right now and whether a limit is reached, and each usage window (on paid plans usually the 5-hour and the weekly window; on Free a single 30-day window) with its percent used, its length in seconds and its reset time (ISO), plus the code-review limit, any extra per-feature limits, and the credits state. ChatGPT chats have no usage meter of their own; these are the Codex limits that come with the ChatGPT plan. Read-only. A signed-out browser fails with an error that says to sign in; a refused or rate-limited usage request fails within seconds with the HTTP status.
Get ChatGPT memory setting
Report whether memory is turned on for the signed-in ChatGPT account: the "Enable memory" switch in Settings > Personalization, which lets ChatGPT use saved memories and past chats to personalize its answers. Requires a signed-in session; a signed-out browser fails with an error that says so. Use set_memory to change it.
OpenAI (ChatGPT) login (Google SSO)
Log into OpenAI / ChatGPT via Google SSO — picks the right account when several are signed in, and does nothing if you're already logged in. If Google interrupts the flow with a password re-confirmation or a 2FA step-up, the script can't complete that step on its own; sign in once manually in the same browser to clear the step-up, then re-run. It may also occasionally fail to complete a fresh login due to the site's own anti-automation checks — if that happens, try again shortly.
Ask ChatGPT
Open a new ChatGPT chat, send one question, wait for the answer to finish (up to 3 minutes), and return the answer, the sources it cited, and the web searches it ran. `answers` is a list: ChatGPT sometimes answers twice and asks which reply you prefer, so check its length before reading answers[0]. `references` are the pages the answer cited, each with the id of the search result it came from when ChatGPT gives one (the search call and the result's rank in it). `sources` are the pages its searches gave the model. `searches` and `calls` also list the results ChatGPT shows under Sources but never gave the model, with `given: false`. An empty `references` means that reply cited nothing: the same question may search on one run and not the next, so repeat it before drawing conclusions. `webSearchQueries` lists what ChatGPT searched for, `searches` groups those queries by round with their results, and `calls` lists each search call with every result's type, position, snippet and date. `activity` gives the progress lines the page showed while it worked ("Searching the web", "Searched 8 websites"). `model`, `stopReason` and `searchContext` say which model answered and how; signed in, `searchContext` also gives the model and thinking effort the account asked for, and its plan. Signed in, it answers in Chat mode only: when the browser is in Work mode it fails with an error that says so; switch back with set_mode (a Chat choice lasts about four hours). Works signed in or signed out, on a Reduck cloud browser or your own paired browser, in any display language. Signed in, every field is filled, and the account's saved memory can influence the answer. Signed out, ChatGPT shows the answer and its citations but not its searches: `answers`, `references` (with each citation's search result id) and `activity` are filled, and `webSearchQueries`, `searches`, `sources`, `calls`, `model` and `stopReason` are null (unknown, not none).
ChatGPT billing: list your ChatGPT subscription invoices
ChatGPT subscription invoices, not the OpenAI API platform. Lists the billing history of the ChatGPT account currently open in this browser, as its Billing page shows it: only that account's invoices, never those of the user's other accounts. To get another account's invoices, switch ChatGPT to that account, then run this again. A Business, Team or Enterprise workspace open in the browser has no personal Billing page, and the run fails saying so. Each invoice comes with its date, amount, currency, payment status, plan and a link to the invoice; the result also names the active account and lists every account the user belongs to. Amounts are in the currency's smallest unit, so 1917 means 19.17 EUR. The Billing page shows the most recent invoices first and this script cannot page further: when ChatGPT says older invoices exist, complete is false and next_cursor holds the cursor ChatGPT gave, but those older invoices are not returned. When the browser is not signed in to ChatGPT, the run fails with a not-signed-in error. Pass hosted_invoice_url to the Stripe invoice PDF download script to get the PDF.
List OpenAI platform invoices
List the invoices of an OpenAI API platform billing account (API usage and credits; for the ChatGPT subscription use chatgpt.com/list_invoices). Returns id, number, total (in cents), amount paid, currency, status, creation time, hosted invoice link and PDF link. Covers the past 12 months, as the billing page does. An account in several organizations picks the first one offered, or the one named exactly by organization; an unknown name fails and lists the organizations offered.
Get Le Chat conversation messages
Full message transcript of one Le Chat conversation by id, oldest to newest.
Ask Le Chat
Open a new Le Chat conversation, send one question, wait for the answer, and return the answer text plus any web-search source domains cited.
Get Gemini chat messages
Full message transcript of one Gemini conversation by id, oldest to newest.
List your X (Twitter) DM inbox
List the conversations in your X chat inbox, in the order X shows them. Returns per conversation: conversation_id, name, handle, last_message, time, sent_by_me (null in a group conversation) and unread. An empty inbox returns an empty list. If X asks for your chat passcode on this browser, pass the 4-digit passcode as pin; without it the run fails with an error asking for it. When the browser is signed out of X, the run fails with a signed-out error asking you to sign in to x.com.
Get Telegram conversation
Open one Telegram conversation by its peer id (the id get_inbox returns) and read its messages in chronological order: message id, text, the time shown on the bubble, whether it carries media, and — for channel posts — the exact view and forward counts. Scrolls upward to load older messages up to the requested count. Returns an empty list for a conversation with no messages, which is a real answer rather than a failure.
Send Chat Message
Send a message in a 1:1 Teams chat by contact name — finds the person, opens their chat, sends the message, then reads the sent message back to confirm it posted.
List Le Chat conversations
List recent Le Chat conversations, newest first.
WhatsApp — Archive chat
Archive or unarchive a chat by exact name. Archived chats still appear in search. Safe to repeat: returns success if the chat is already in the desired state.
Get LinkedIn conversation attendees
List the participants of a classic LinkedIn messaging conversation (NOT Sales Navigator) by thread URL from list_inbox — mainly useful for group chats. Returns each attendee's name, profile URL and member urn, plus whether the thread is a group.
Get Claude chat messages
Full message transcript of one Claude.ai conversation by uuid, oldest→newest. `text` is the readable prose; `tools` lists tool calls the UI renders as cards.
Le Chat (Mistral) login (Google SSO)
Log into Le Chat (chat.mistral.ai) with Google sign-in, handling the Google account chooser and consent screen. Returns straight away if this browser is already signed in, and reports the Mistral account email. Optionally pass which Google account to use. Needs a browser already signed in to Google and a Google account that already has a Mistral account.
Search Telegram messages in a conversation
Search inside one Telegram conversation using the conversation's own search box, and return the matching messages: the chat they are in, the date label shown on the result, and the matched text. Returns an empty list when nothing matches, which is a real answer. Note the result rows carry no message id, so results cannot be fed straight to delete or react — use get_conversation for ids. Searching across all chats is a separate Telegram surface and is not covered here.
Delete Le Chat conversation
Delete one Le Chat conversation by id (moves it to trash; Le Chat purges trashed conversations automatically after 30 days).
Search Gemini chats
Search your Gemini conversations (the "Search chats" box): server-side full-text over titles and message bodies.