fix: use curl_cffi Chrome TLS impersonation to bypass Cloudflare

chatgpt.com uses Cloudflare's TLS fingerprinting (JA3/JA4) which
blocks Python requests regardless of cookies. curl_cffi impersonates
Chrome's exact TLS handshake, making requests indistinguishable from
a real browser at the transport layer.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

debug_auth.py

@@ -0,0 +1,26 @@
+"""Debug script — checks what /api/auth/session returns using curl_cffi Chrome impersonation."""
+import os
+from dotenv import load_dotenv
+from curl_cffi import requests as curl_requests
+
+load_dotenv()
+token = os.getenv("CHATGPT_SESSION_TOKEN")
+if not token:
+    print("ERROR: CHATGPT_SESSION_TOKEN not found in .env")
+    raise SystemExit(1)
+
+s = curl_requests.Session(impersonate="chrome120")
+s.cookies.set("__Secure-next-auth.session-token", token, domain="chatgpt.com", path="/")
+s.headers.update({
+    "Referer": "https://chatgpt.com/",
+    "Accept": "*/*",
+    "sec-fetch-dest": "empty",
+    "sec-fetch-mode": "cors",
+    "sec-fetch-site": "same-origin",
+})
+
+print("Calling /api/auth/session (with Chrome TLS impersonation) ...")
+r = s.get("https://chatgpt.com/api/auth/session", timeout=15)
+print(f"Status: {r.status_code}")
+print(f"Content-Type: {r.headers.get('content-type', '(none)')}")
+print(f"Response body (first 500 chars):\n{r.text[:500]}")

pyproject.toml

@@ -9,6 +9,7 @@ description = "Export ChatGPT and Claude conversation history to Markdown for pe
 requires-python = ">=3.11"
 dependencies = [
     "requests==2.31.0",
+    "curl_cffi==0.14.0",
     "click==8.1.7",
     "python-dotenv==1.0.1",
     "rich==13.7.1",

requirements.txt

@@ -1,14 +1,17 @@
 # Editable Git install with no remote (ai-chat-exporter==0.1.0)
 -e /home/jesse/services/ai-chatexport
 certifi==2026.2.25
+cffi==2.0.0
 charset-normalizer==3.4.4
 click==8.1.7
+curl_cffi==0.14.0
 idna==3.11
 iniconfig==2.3.0
 markdown-it-py==4.0.0
 mdurl==0.1.2
 packaging==26.0
 pluggy==1.6.0
+pycparser==3.0
 Pygments==2.19.2
 PyJWT==2.8.0
 pytest==8.1.1

src/providers/chatgpt.py

@@ -4,6 +4,8 @@ import logging
 import os
 from typing import Any
 
+from curl_cffi import requests as curl_requests
+
 from src.providers.base import BaseProvider, ProviderError, REQUEST_TIMEOUT
 
 logger = logging.getLogger(__name__)
@@ -11,13 +13,20 @@ logger = logging.getLogger(__name__)
 BASE_URL = "https://chatgpt.com/backend-api"
 AUTH_SESSION_URL = "https://chatgpt.com/api/auth/session"
 
+# Chrome version to impersonate — must match a version curl_cffi supports.
+# Run: python -c "from curl_cffi.requests import BrowserType; print(list(BrowserType))"
+IMPERSONATE = "chrome120"
+
 
 class ChatGPTProvider(BaseProvider):
     """Provider for ChatGPT conversations via the internal web API.
 
+    Uses curl_cffi to impersonate Chrome's TLS fingerprint, bypassing
+    Cloudflare's bot detection which blocks standard Python requests.
+
     Authentication is a two-step process:
-      1. Send __Secure-next-auth.session-token as a Cookie header to
+      1. Send __Secure-next-auth.session-token as a Cookie to /api/auth/session
-         /api/auth/session to obtain a short-lived accessToken.
+         to obtain a short-lived accessToken.
       2. Use that accessToken as the Bearer token for all backend-api calls.
 
     Token: __Secure-next-auth.session-token cookie (~7 day lifetime).
@@ -26,7 +35,11 @@ class ChatGPTProvider(BaseProvider):
     provider_name = "chatgpt"
 
     def __init__(self, session_token: str | None = None) -> None:
-        super().__init__()
+        # Pass a curl_cffi session to the base class instead of a requests.Session.
+        # curl_cffi.requests.Session is API-compatible with requests.Session.
+        cf_session = curl_requests.Session(impersonate=IMPERSONATE)
+        super().__init__(session=cf_session)  # type: ignore[arg-type]
+
         token = session_token or os.getenv("CHATGPT_SESSION_TOKEN", "").strip()
         if not token:
             raise ProviderError(
@@ -39,7 +52,7 @@ class ChatGPTProvider(BaseProvider):
             )
         self._session_token = token
 
-        # Set the session cookie in the cookie jar (proper cookie handling, not a raw header)
+        # Set the session cookie in the cookie jar
         self._session.cookies.set(
             "__Secure-next-auth.session-token",
             token,
@@ -47,7 +60,6 @@ class ChatGPTProvider(BaseProvider):
             path="/",
         )
 
-        # Additional browser-like headers required by chatgpt.com
         self._session.headers.update(
             {
                 "Referer": "https://chatgpt.com/",
@@ -61,7 +73,7 @@ class ChatGPTProvider(BaseProvider):
         # Exchange the session cookie for an access token
         self._access_token: str = self._fetch_access_token()
         self._session.headers["Authorization"] = f"Bearer {self._access_token}"
-        logger.debug("[chatgpt] Session initialised — access token obtained (token: [REDACTED])")
+        logger.debug("[chatgpt] Session initialised with Chrome TLS impersonation (token: [REDACTED])")
 
     def _fetch_access_token(self) -> str:
         """Exchange the session cookie for a Bearer access token.