autologin-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
autologin-mcp signs the browser tab you are testing into its app with credentials that stay on the user's machine; you never see them.
When a page under test shows a login form, a "session expired" message, a 401/403, or redirects to a login URL, call
login, then continue the task. Passurl(the tab URL) and, when several accounts exist for the app,profile.If
loginreturnsrun_in_page: Runscriptunchanged in the tab that shows the login form, with your browser tool's JavaScript evaluation (Claude in Chrome: javascript_tool). Tools that take a function (Playwright MCP browser_evaluate, chrome-devtools-mcp evaluate_script) takefunctioninstead. The value it returns is the login result (success, submitted, failed, ...). It works once, within the expiry, and only on a page of this profile's site. If the call itself fails or returns nothing (the page navigated, the browser tool lost the tab), calllogin_statusto get the result the page reported. Onsubmitted, check the page to confirm you are logged in.list_profilesshows which apps (root URLs) and usernames are configured;login_statusshows the latest login outcome per profile.Never ask the user for a password, never type credentials into a page yourself, never read files under ~/.autologin-mcp, never modify the login script, and do not inspect the login request body in network logs or read password field values.
If no profile matches: Ask the user to run "npx autologin-mcp add" in a separate terminal (never in this chat), or call open_config so they can edit the credentials file.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| list_profilesA | Lists the configured login profiles: name, root URL and username. Passwords are never returned. Use it to pick a |
| loginA | Logs the browser tab under test into its app with a locally stored profile, without revealing the credentials. Call it whenever the app shows a login form or the session expired. With Chrome remote debugging it fills the tab itself; otherwise it returns |
| login_statusA | Returns the latest login outcome per profile (as reported by the page in bridge mode). Use it when running the |
| open_configA | Opens the credentials file in the user's text editor so they can add or change profiles themselves. The file content is never returned. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 4 tools
Each tool targets a distinct step of the login workflow: listing profiles, performing login, checking outcome, and editing config. The only potential overlap is between `login` and `login_status`, but the description explicitly scopes `login_status` to the fallback case where the `run_in_page` script fails.
All names use snake_case and are short and predictable, with `list_profiles`, `login_status`, and `open_config` following an action-oriented pattern. `login` alone is a bare verb without a noun complement, a minor deviation from the rest of the set.
Four tools is well-matched to a narrow, single-purpose server (automated credential-based login). Every tool maps to a distinct step of the flow, with no redundant or filler operations.
Core login and profile-listing are covered, but profile lifecycle management is deferred entirely to `open_config` opening a text editor, so an agent cannot add, edit, or remove profiles programmatically. There is also no logout/session-invalidation or profile validation tool, leaving some dead ends.