How to Configure Twitter Authentication Cookies for Agent-Reach: A Complete Guide

Agent-Reach reads Twitter authentication data from ~/.agent-reach/config.yaml and supports both manual CLI entry and automatic extraction from Chrome, Firefox, Edge, Brave, or Opera to authenticate with the Twitter API.

Agent-Reach is an open-source automation framework by Panniantong that enables agent-based interactions through multiple channels, including Twitter. To post or retrieve data via the TwitterChannel implementation, you must configure Twitter authentication cookies so the system can authenticate through twitter-cli or the OpenCLI backend. This guide covers the two supported configuration methods, how credentials are persisted, and how to verify they function correctly.

Manual Configuration via CLI

The fastest way to supply credentials is using the built-in CLI configuration command. Agent-Reach accepts the two required cookies—auth_token and ct0—either as a formatted string or as separate arguments.

Using the Cookie String Format

Pass the full cookie header value directly to the configure command. The CLI parses this input in _parse_twitter_cookie_input within agent_reach/cli.py (lines 54-73), which splits the string and stores the values under twitter_auth_token and twitter_ct0 in the configuration file.

agent-reach configure twitter-cookies "auth_token=AAA; ct0=BBB"

Using Separate Arguments

Alternatively, provide the two tokens as space-separated values without key names:

agent-reach configure twitter-cookies AAA BBB

Both methods write the credentials to ~/.agent-reach/config.yaml and immediately run a health check against twitter-cli to confirm they are valid.

Automatic Extraction from Browser

If you are already logged into Twitter (x.com) in a desktop browser, Agent-Reach can extract the cookies automatically. This method is implemented in agent_reach/cookie_extract.py and supports Chrome, Firefox, Edge, Brave, and Opera.

Run the following command, replacing chrome with your browser name:

agent-reach configure --from-browser chrome

The handler in agent_reach/cli.py (_cmd_configure, lines 26-34) delegates to configure_from_browser, which scans the selected browser’s cookie store for domains .x.com and .twitter.com. It filters for the required auth_token and ct0 cookies (lines 14-22) and writes them back via config.set("twitter_auth_token", ...) and config.set("twitter_ct0", ...) (lines 50-58).

How the Credentials Are Stored and Used

Once configured, Agent-Reach persists the cookies in ~/.agent-reach/config.yaml as defined in agent_reach/config.py (lines 24-27). The system synchronizes these credentials across several downstream tools to ensure compatibility:

• twitter-cli: The CLI sets environment variables TWITTER_AUTH_TOKEN and TWITTER_CT0 during the health-check phase so twitter-cli can read them immediately.
• xfetch compatibility: The legacy layer writes the same pair into ~/.config/xfetch/session.json (lines 76-84 in cookie_extract.py).
• bird CLI: Agent-Reach can generate a credentials.env file for the bird tool (lines 104-112 in cookie_extract.py).

Verifying Your Configuration

After configuration, verify that the Twitter channel is operational using the built-in diagnostic command:

agent-reach doctor

According to the source in agent_reach/channels/twitter.py (lines 59-67), a successful setup reports "Twitter CLI 完整可用" (Twitter CLI fully available) or "OpenCLI 可用" (OpenCLI available).

For manual verification, export the environment variables and run a status check:

export TWITTER_AUTH_TOKEN=AAA
export TWITTER_CT0=BBB
twitter status

If the output contains ok: true, your cookies are valid and the authentication flow is complete.

Summary

• Agent-Reach requires two cookies—auth_token and ct0—to authenticate with Twitter, stored in ~/.agent-reach/config.yaml.
• Manual configuration uses agent-reach configure twitter-cookies with either a formatted string or separate values, parsed by _parse_twitter_cookie_input in cli.py.
• Automatic extraction pulls cookies from Chrome, Firefox, Edge, Brave, or Opera using agent-reach configure --from-browser <browser>.
• Credentials are synchronized to environment variables and legacy files like ~/.config/xfetch/session.json.
• Verify the setup with agent-reach doctor or by running twitter status after exporting TWITTER_AUTH_TOKEN and TWITTER_CT0.

Frequently Asked Questions

Where does Agent-Reach store Twitter cookies?

Agent-Reach stores the auth_token and ct0 values in ~/.agent-reach/config.yaml under the keys twitter_auth_token and twitter_ct0, as implemented in agent_reach/config.py (lines 24-27).

Which browsers are supported for automatic cookie extraction?

The automatic extraction feature in agent_reach/cookie_extract.py supports Chrome, Firefox, Edge, Brave, and Opera. It scans the browser’s cookie store for domains .x.com and .twitter.com to retrieve the required tokens.

How do I verify that my Twitter authentication is working?

Run agent-reach doctor and check the Twitter section for the message "Twitter CLI 完整可用" or "OpenCLI 可用". Alternatively, manually verify by exporting TWITTER_AUTH_TOKEN and TWITTER_CT0 and running twitter status to ensure it returns ok: true.

Can I use the cookies with other Twitter CLI tools?

Yes. Agent-Reach synchronizes credentials to multiple formats: it sets environment variables for twitter-cli, writes to ~/.config/xfetch/session.json for xfetch compatibility, and can generate a credentials.env file for the bird CLI, as defined in agent_reach/cookie_extract.py (lines 76-112).

{
  "mcpServers": {
    "instagit": {
      "command": "npx",
      "args": ["-y", "instagit@latest"]
    }
  }
}