How to Configure Twitter/X Cookies for Agent-Reach: Manual and Automatic Methods

Agent-Reach requires the auth_token and ct0 cookies from your X (Twitter) session to enable search functionality, which you can configure either manually via the CLI or extract automatically from your browser.

The open-source Panniantong/Agent-Reach repository provides automation tools for social media platforms, but accessing Twitter/X data requires valid authentication cookies stored in your local configuration. This guide explains exactly how to configure Twitter cookies for Agent-Reach using the two supported methods: manual CLI input and automatic browser extraction.

Required Cookies for X Authentication

Agent-Reach specifically needs two authentication tokens that X uses to maintain your session: auth_token and ct0. These values are persisted in your user configuration file at ~/.agent-reach/config.yaml under the keys twitter_auth_token and twitter_ct0.

According to the source code in agent_reach/config.py, the Config class manages these values using the set() method to store sensitive credentials securely. When properly configured, these cookies enable the TwitterChannel class (defined in agent_reach/channels/twitter.py) to authenticate requests and perform search operations on your behalf.

Method 1: Manual Configuration via CLI

The most direct way to configure Twitter cookies for Agent-Reach is using the configure twitter-cookies command, which accepts either separate arguments or a raw cookie header string.

Pass the values as separate arguments:

agent-reach configure twitter-cookies AUTH_TOKEN_VALUE CT0_VALUE

Or paste a raw cookie header (copied directly from Chrome’s “Copy > Copy Cookies”):

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

The CLI parses your input using the _parse_twitter_cookie_input function in agent_reach/cli.py (lines 24-42) to extract the relevant tokens regardless of extra cookie data. After storage, the tool validates the credentials by invoking twitter status via subprocess and checking for "ok: true" in the output. Success appears as:

✅ Twitter cookies configured!
Testing Twitter access... ✅ Twitter access works!

Method 2: Automatic Extraction from Browser

Agent-Reach can automatically extract auth_token and ct0 from your existing browser sessions without manual copy-pasting. This method supports Chrome, Firefox, Edge, Brave, and Opera.

Run the configuration command with the browser flag:

agent-reach configure --from-browser chrome

Under the hood, the configure_from_browser function in agent_reach/cookie_extract.py iterates through supported platforms and extracts cookies using either rookiepy or browser-cookie3. When it finds the Twitter cookies, it stores them via:

config.set("twitter_auth_token", tc["auth_token"])
config.set("twitter_ct0", tc["ct0"])

The function also synchronizes these credentials to legacy helper tools through _sync_xfetch_session (writing to ~/.config/xfetch/session.json) and _sync_bird_env (writing to ~/.config/bird/credentials.env). Successful extraction produces output similar to:

Extracting cookies from chrome...

✅ Twitter/X: auth_token + ct0
...
✅ Cookies configured! Run `agent-reach doctor` to see updated status.

Where Credentials Are Stored and Synced

Your Twitter cookies are stored in multiple locations to ensure compatibility with different Agent-Reach components:

• Primary storage: ~/.agent-reach/config.yaml contains twitter_auth_token and twitter_ct0 as defined in agent_reach/config.py
• xfetch legacy: ~/.config/xfetch/session.json receives synced values via _sync_xfetch_session for backward compatibility
• bird-cli: ~/.config/bird/credentials.env receives a source-able file via _sync_bird_env containing AUTH_TOKEN and CT0 environment variables

You can verify the stored configuration (with redacted values) by viewing the YAML file directly:

cat ~/.agent-reach/config.yaml

Troubleshooting Validation Failures

If you see an error during the "Testing Twitter access" phase, the cookies may be expired or invalid. The validation logic in agent_reach/cli.py (within _cmd_configure, lines 65-78) checks the subprocess output of twitter status and reports specific errors if "ok: true" is missing.

To resolve validation failures:

1. Re-run the configuration with fresh cookies from your browser
2. Ensure you are using the --from-browser flag if manually copied values fail
3. Check that twitter-cli is installed and accessible in your PATH, as Agent-Reach relies on this external tool to verify credentials

Summary

• Agent-Reach requires two specific cookies: auth_token and ct0 from X (Twitter) to enable search capabilities.
• Configuration file: Values are stored in ~/.agent-reach/config.yaml under twitter_auth_token and twitter_ct0.
• Two configuration methods: Manual CLI input using agent-reach configure twitter-cookies or automatic extraction via agent-reach configure --from-browser chrome.
• Legacy synchronization: The tool automatically syncs credentials to ~/.config/xfetch/session.json and ~/.config/bird/credentials.env for helper tool compatibility.
• Validation: The CLI runs twitter status after configuration to verify connectivity and reports immediate success or failure.

Frequently Asked Questions

What are the auth_token and ct0 cookies?

The auth_token is your primary session identifier for X (formerly Twitter), while ct0 is a cross-site request forgery (CSRF) token required for API calls. According to the Agent-Reach source code, both must be present in ~/.agent-reach/config.yaml for the TwitterChannel to authenticate requests successfully.

Which browsers support automatic cookie extraction?

Agent-Reach supports automatic extraction from Chrome, Firefox, Edge, Brave, and Opera. The configure_from_browser function in agent_reach/cookie_extract.py uses platform-specific libraries (rookiepy or browser-cookie3) to read the SQLite cookie databases from these browsers without requiring manual export.

Where does Agent-Reach store my Twitter credentials?

Credentials are primarily stored in ~/.agent-reach/config.yaml using the Config class from agent_reach/config.py. Additionally, the tool synchronizes these values to legacy locations: ~/.config/xfetch/session.json for the xfetch tool and ~/.config/bird/credentials.env for bird-cli compatibility.

Why does the configuration fail validation even with correct cookies?

Validation failures usually occur when the external twitter-cli tool is not installed in your PATH, or when the cookies have expired. The _cmd_configure function in agent_reach/cli.py runs twitter status via subprocess and checks for "ok: true" in the output. Ensure your shell can execute twitter status independently, or re-extract fresh cookies if your session expired.

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