credentials_list

Lists the project's saved credentials, passwords and secrets. For each one it shows the name, the type, the sites it may be typed on, where it may be used and, for a password, the parts browser_type can type and where its 2FA codes come from. Use a name with browser_type's credential input, with browser_login, or in session_create's credentials.

Inputs

None.

Example call

This is what an MCP client sends. You do not write it yourself: ask the model in plain words and it calls the tool.

{  "name": "credentials_list",  "arguments": {}}

What it returns

A JSON list with one entry per credential, or a line saying that none are saved.

[  {    "name": "SHOP",    "type": "password",    "sites": ["https://shop.example.com"],    "scope": "all",    "fields": ["username", "password", "otp"],    "codeSource": "totp",    "browserType": true  },  {    "name": "GITHUB_TOKEN",    "type": "secret",    "sites": "any",    "scope": "shell",    "browserType": false  }]

When a call fails, the result is marked as an error. Its text starts with the API’s error code, such as unauthorized or session_not_running, followed by its message.

Annotations

The server tells every client what this tool does to the world, so a client can decide what to ask you about before it runs.

Good to know

• It never shows a value. Credentials are added in the console, with the CLI or with an SDK: no tool creates or changes one, so a password never passes through the chat.
• browserType is false for a credential with scope shell: browser_type is refused for it.
• otp is in fields only for a password that has a 2FA code source.
• codeSource says where a password's 2FA codes come from: totp (an authenticator key), push or url (your own system sends each code or sign-in link), or null (no 2FA). The address a url source asks, and its signing secret, are never shown.
• Not available when the server runs with MCP_TOOLS=directory.

The same thing with the API