From c32c7fb95999999fa3f3e108eeb061bf2ca6b6d9 Mon Sep 17 00:00:00 2001
From: Otto <otto@agpt.co>
Date: Thu, 5 Feb 2026 19:11:56 +0000
Subject: [PATCH] fix: Handle RFC 2045 line-wrapped base64 in binary detection

- Strip whitespace before base64 decoding (encoders often add line breaks)
- Update regex comment to clarify whitespace handling
- Add test case for line-wrapped base64

Addresses CodeRabbit review feedback.
---
 .../features/chat/tools/binary_output_processor.py | 10 +++++++---
 .../chat/tools/test_binary_output_processor.py     | 14 ++++++++++++++
 2 files changed, 21 insertions(+), 3 deletions(-)

diff --git a/autogpt_platform/backend/backend/api/features/chat/tools/binary_output_processor.py b/autogpt_platform/backend/backend/api/features/chat/tools/binary_output_processor.py
index 9ac0812b0f..c2c246534c 100644
--- a/autogpt_platform/backend/backend/api/features/chat/tools/binary_output_processor.py
+++ b/autogpt_platform/backend/backend/api/features/chat/tools/binary_output_processor.py
@@ -47,7 +47,8 @@ ALLOWED_MIMETYPES = {
 }
 
 # Base64 character validation (strict - must be pure base64)
-BASE64_PATTERN = re.compile(r"^[A-Za-z0-9+/\n\r]+=*$")
+# Allows whitespace which will be stripped before decoding (RFC 2045 line wrapping)
+BASE64_PATTERN = re.compile(r"^[A-Za-z0-9+/\s]+=*$")
 
 # Magic numbers for binary file detection
 # Note: WebP requires two-step detection: RIFF prefix + WEBP at offset 8
@@ -171,12 +172,15 @@ def _detect_raw_base64(value: str) -> Optional[tuple[bytes, str]]:
 
     Returns (content, extension) or None.
     """
-    # Pre-filter: must look like base64 (no spaces, punctuation, etc.)
+    # Pre-filter: must look like base64 (allows whitespace for RFC 2045 line wrapping)
     if not BASE64_PATTERN.match(value):
         return None
 
+    # Strip whitespace before decoding (RFC 2045 allows line breaks in base64)
+    normalized = re.sub(r"\s+", "", value)
+
     try:
-        content = base64.b64decode(value, validate=True)
+        content = base64.b64decode(normalized, validate=True)
     except (ValueError, binascii.Error):
         return None
 
diff --git a/autogpt_platform/backend/backend/api/features/chat/tools/test_binary_output_processor.py b/autogpt_platform/backend/backend/api/features/chat/tools/test_binary_output_processor.py
index b6f54a3a17..312fafa4be 100644
--- a/autogpt_platform/backend/backend/api/features/chat/tools/test_binary_output_processor.py
+++ b/autogpt_platform/backend/backend/api/features/chat/tools/test_binary_output_processor.py
@@ -198,6 +198,20 @@ class TestDetectRawBase64:
         result = _detect_raw_base64("not-valid-base64!!!")
         assert result is None
 
+    def test_detects_base64_with_line_breaks(self):
+        """Should detect raw base64 with RFC 2045 line breaks."""
+        png_content = b"\x89PNG\r\n\x1a\n" + b"\x00" * 100
+        png_b64 = base64.b64encode(png_content).decode()
+        # Simulate RFC 2045 line wrapping at 76 chars
+        wrapped = png_b64[:76] + "\n" + png_b64[76:]
+
+        result = _detect_raw_base64(wrapped)
+
+        assert result is not None
+        content, ext = result
+        assert ext == "png"
+        assert content == png_content
+
 
 # =============================================================================
 # Process Binary Outputs Tests