gmail without an api key

    1

    Control Gmail by driving a signed-in Chrome session via CDP, bypassing OAuth and API key requirements.

    $29

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    gmail without an api key

    Example session with this skill installed

    Search my Gmail for recent invoices from 'acme.com' and show me the last two subjects.

    • Read your context and instructions
    • Compiled the gmail without an

    I have searched your Gmail for messages matching the query from:acme.com invoice and retrieved the two most recent threads.

    Gmail Search Results

    DateFromSubjectSnippet
    Oct 24Acme Corp BillingInvoice #INV-2024-001Your subscription for October has been processed...
    Sep 24Acme Corp BillingInvoice #INV-2024-098Please find the attached invoice for last month...

    Next steps

    1. Run node scripts/gmail.mjs open --index 0 to read the full body of the most recent invoice.
    2. Use the compose command if you need to draft a reply or forward these details.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    The Gmail API needs a Cloud project, an OAuth client, a consent screen, a refresh token and quota management. Most agent jobs need none of that.

    This skill drives the Gmail page that is already open, on an account that is already signed in. The browser session is the authentication.

    It ships:

    • a zero-dependency CDP client (scripts/cdp.mjs, Node 22+) - a Gmail CLI (scripts/gmail.mjs): list the inbox, search, open a message and read its real headers, and compose with a read-back of the recipient chips before anything is sent - --send is explicit, and the CLI refuses to send when no recipient chip is present

    The SKILL.md documents the traps: the Sent folder shows the sender and reads like a self-send, an empty To field after typing is the proof of success, recipient queries must be scoped to the compose dialog, hidden tabs return empty innerText, and a mailbox may have no signature at all.

    Read freely. Send deliberately. Nothing is sent without an explicit flag.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Workflow Automation.

    No support for attachments, labels, or thread replies. Requires manual Chrome flag configuration. Only works with already signed-in sessions.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before buying a skill.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 27 days ago

    Permissions

    Terminal / Shell
    Listed27 days ago

    What's inside

    Frequently Asked Questions

    This skill uses the Chrome DevTools Protocol (CDP) to interact with a signed-in browser session, allowing you to bypass complex Google Cloud API setups, OAuth consent screens, and token management.

    You must launch Chrome with the `--remote-debugging-port` flag enabled and be signed into the desired Gmail account; the skill then connects to that active session to perform actions.

    The skill includes a specialized Node.js driver and scripts for listing, searching, reading, and composing emails, ensuring reliable DOM interaction without manual selector configuration.

    Yes, the skill includes logic to verify account slots (like /u/0/ or /u/1/), allowing you to specify which signed-in mailbox the agent should interact with.

    While it handles text-based drafting and sending, it currently does not support file attachments, managing labels, or replying directly to existing threads.

    The skill uses specific verification logic to ensure 'recipient chips' are correctly rendered in the Gmail UI before sending, preventing common agent errors like sending to unvalidated addresses.

    New in Workflow Automation