hermes-agent/agent/image_routing.py
Shashwat Gokhe 505bc27d8d fix(gateway): classify mixed attachments per-attachment + transcode uncommon image formats
A document attached alongside an image in the same Discord message was
swept into the vision pipeline and 400'd the whole turn ("Could not
process image"), and was simultaneously never surfaced to the agent as a
readable file. Restores the "any file type works" contract for mixed
messages and fixes the HTTP 400.

Bug 1 — mixed attachments: the inbound routing loop keyed image/audio/video
classification off the message-level type (PHOTO/VOICE/AUDIO), so a doc in
a PHOTO message landed in image_paths and poisoned the vision call. The
document context-note path was gated on message_type == DOCUMENT, so that
same doc never reached the agent at all. Now classification is
per-attachment (trust each attachment's own MIME; fall back to the
message-level type only when MIME is unknown), via shared _event_media_is_*
helpers used by both _build_media_placeholder and the main inbound loop.
The document note now fires for any non-image/audio/video attachment
regardless of message-level type.

Bug 2 — uncommon formats: AVIF/HEIC/BMP/TIFF/ICO produced the same generic
400 because providers only accept PNG/JPEG/GIF/WEBP. image_routing now
transcodes those to PNG via Pillow before declaring media_type, skipping
cleanly (logged) if Pillow/plugins are missing. SVG is vector — Pillow
can't rasterize it — so it's skipped rather than transcoded.

Closes #25935.

Co-authored-by: LeonSGP43 <cine.dreamer.one@gmail.com>
Co-authored-by: cypres0099 <74935762+cypres0099@users.noreply.github.com>
2026-06-27 19:26:04 -07:00

650 lines
25 KiB
Python

"""Routing helpers for inbound user-attached images.
Two modes:
native — attach images as OpenAI-style ``image_url`` content parts on the
user turn. Provider adapters (Anthropic, Gemini, Bedrock, Codex,
OpenAI chat.completions) already translate these into their
vendor-specific multimodal formats.
text — run ``vision_analyze`` on each image up-front and prepend the
description to the user's text. The model never sees the pixels;
it only sees a lossy text summary. This is the pre-existing
behaviour and still the right choice for non-vision models.
The decision is made once per message turn by :func:`decide_image_input_mode`.
It reads ``agent.image_input_mode`` from config.yaml (``auto`` | ``native``
| ``text``, default ``auto``) and the active model's capability metadata.
In ``auto`` mode:
- If the user has explicitly configured ``auxiliary.vision.provider``
(i.e. not ``auto`` and not empty), we assume they want the text pipeline
regardless of the main model — they've opted in to a specific vision
backend for a reason (cost, quality, local-only, etc.).
- Otherwise, if the active model reports ``supports_vision=True`` in its
models.dev metadata, we attach natively.
- Otherwise (non-vision model, no explicit override), we fall back to text.
This keeps ``vision_analyze`` surfaced as a tool in every session — skills
and agent flows that chain it (browser screenshots, deeper inspection of
URL-referenced images, style-gating loops) keep working. The routing only
affects *how user-attached images on the current turn* are presented to the
main model.
"""
from __future__ import annotations
import base64
import logging
import mimetypes
import os
import re
from pathlib import Path
from typing import Any, Dict, List, Optional, Tuple
logger = logging.getLogger(__name__)
_VALID_MODES = frozenset({"auto", "native", "text"})
# Image extensions used by extract_image_refs(). Kept tight on purpose — we
# only auto-attach things the model can actually see. Documents/archives are
# excluded because the gateway's broader extract_local_files() also routes
# them differently (send_document), and we don't want to attach a PDF as a
# vision part.
_IMAGE_EXTS = (
".png", ".jpg", ".jpeg", ".gif", ".webp", ".bmp", ".tiff", ".tif", ".heic",
)
_IMAGE_EXT_PATTERN = "|".join(e.lstrip(".") for e in _IMAGE_EXTS)
# Absolute / home-relative local image path. Matches the same shape gateway's
# extract_local_files() uses: anchors to ``~/`` or ``/``, ignores matches inside
# URLs (the ``(?<![/:\w.])`` lookbehind), and case-insensitive on the extension.
_LOCAL_IMAGE_PATH_RE = re.compile(
r"(?<![/:\w.])(?:~/|/)(?:[\w.\-]+/)*[\w.\-]+\.(?:" + _IMAGE_EXT_PATTERN + r")\b",
re.IGNORECASE,
)
# http(s) URL ending in an image extension (optionally followed by a
# query string). Case-insensitive on the extension. Strict ``http(s)://``
# scheme so we don't accidentally grab ``file://`` URLs or other shapes.
_IMAGE_URL_RE = re.compile(
r"https?://[^\s<>\"']+?\.(?:" + _IMAGE_EXT_PATTERN + r")(?:\?[^\s<>\"']*)?",
re.IGNORECASE,
)
def extract_image_refs(text: str) -> Tuple[List[str], List[str]]:
"""Scan free-form text for image references the model should see.
Returns ``(local_paths, urls)``:
* ``local_paths`` — absolute (``/``) or home-relative (``~/``) paths
whose suffix is an image extension AND whose expanded form exists
on disk as a file. Order-preserving, deduplicated.
* ``urls`` — ``http(s)://…`` URLs whose path ends in an image
extension (a ``?query`` is allowed after the extension).
Order-preserving, deduplicated.
Matches inside fenced code blocks (``` ``` ```) and inline backticks
(`` `…` ``) are skipped so that snippets pasted into a task body for
reference aren't mistaken for live attachments. This mirrors the
behaviour of ``gateway.platforms.base.BaseAdapter.extract_local_files``.
Local paths are validated against the filesystem; URLs are not
(the provider fetches them at request time).
"""
if not isinstance(text, str) or not text:
return [], []
# Build spans covered by fenced code blocks and inline code so we can
# ignore references the author embedded purely as example text.
code_spans: list[tuple[int, int]] = []
for m in re.finditer(r"```[^\n]*\n.*?```", text, re.DOTALL):
code_spans.append((m.start(), m.end()))
for m in re.finditer(r"`[^`\n]+`", text):
code_spans.append((m.start(), m.end()))
def _in_code(pos: int) -> bool:
return any(s <= pos < e for s, e in code_spans)
local_paths: list[str] = []
seen_paths: set[str] = set()
for match in _LOCAL_IMAGE_PATH_RE.finditer(text):
if _in_code(match.start()):
continue
raw = match.group(0)
expanded = os.path.expanduser(raw)
try:
if not os.path.isfile(expanded):
continue
except OSError:
# ENAMETOOLONG / EINVAL on pathological inputs — skip rather than crash.
continue
if expanded in seen_paths:
continue
seen_paths.add(expanded)
local_paths.append(expanded)
urls: list[str] = []
seen_urls: set[str] = set()
for match in _IMAGE_URL_RE.finditer(text):
if _in_code(match.start()):
continue
url = match.group(0)
# Strip trailing punctuation that's almost certainly prose, not part
# of the URL (e.g. "see https://x.com/a.png." or "/a.png)").
url = url.rstrip(".,;:!?)]>")
if url in seen_urls:
continue
seen_urls.add(url)
urls.append(url)
return local_paths, urls
# Strict YAML/JSON boolean coercion for capability overrides.
#
# ``bool("false")`` is True in Python because non-empty strings are truthy, so
# a user writing ``supports_vision: "false"`` (quoted — a common YAML mistake)
# would silently enable native vision routing on a model that can't actually
# handle it. Accept only the values YAML 1.1 / 1.2 treat as booleans, plus
# real ``bool`` and integer 0/1. Anything else returns None so the caller
# falls through to models.dev rather than honouring garbage.
_TRUE_TOKENS = frozenset({"true", "yes", "on", "1"})
_FALSE_TOKENS = frozenset({"false", "no", "off", "0"})
def _coerce_capability_bool(raw: Any) -> Optional[bool]:
"""Return True/False for recognised boolean values, None otherwise."""
if isinstance(raw, bool):
return raw
if isinstance(raw, int):
if raw in (0, 1):
return bool(raw)
return None
if isinstance(raw, str):
s = raw.strip().lower()
if s in _TRUE_TOKENS:
return True
if s in _FALSE_TOKENS:
return False
return None
def _supports_vision_override(
cfg: Optional[Dict[str, Any]],
provider: str,
model: str,
) -> Optional[bool]:
"""Resolve user-declared vision capability from config.yaml.
Resolution order, first hit wins:
1. ``model.supports_vision`` (top-level shortcut for the active model)
2. ``providers.<provider>.models.<model>.supports_vision``
(named custom providers — ``provider`` may be the runtime-resolved
value ``"custom"`` and/or the user-declared name under
``model.provider``; both are tried)
Returns None when no override is set, so the caller falls through to
models.dev. Returns False explicitly only when the user wrote a
recognised boolean false token.
"""
if not isinstance(cfg, dict):
return None
# 1. Top-level shortcut
model_cfg_raw = cfg.get("model")
model_cfg: Dict[str, Any] = model_cfg_raw if isinstance(model_cfg_raw, dict) else {}
top = _coerce_capability_bool(model_cfg.get("supports_vision"))
if top is not None:
return top
# 2. Per-provider, per-model. Named custom providers (e.g. "my-vllm")
# get rewritten to provider="custom" at runtime
# (hermes_cli/runtime_provider.py:_resolve_named_custom_runtime), so the
# config still holds the user-declared name under model.provider. Try
# both as candidate provider keys.
config_provider = str(model_cfg.get("provider") or "").strip()
providers_raw = cfg.get("providers")
providers_cfg: Dict[str, Any] = providers_raw if isinstance(providers_raw, dict) else {}
for p in dict.fromkeys(filter(None, (provider, config_provider))):
entry_raw = providers_cfg.get(p)
entry: Dict[str, Any] = entry_raw if isinstance(entry_raw, dict) else {}
models_raw = entry.get("models")
models_cfg: Dict[str, Any] = models_raw if isinstance(models_raw, dict) else {}
per_model_raw = models_cfg.get(model)
per_model: Dict[str, Any] = per_model_raw if isinstance(per_model_raw, dict) else {}
coerced = _coerce_capability_bool(per_model.get("supports_vision"))
if coerced is not None:
return coerced
# 2b. Legacy list-style custom_providers. Entries are dicts with a
# "name" key and a nested "models" dict. Match by provider name (which
# may appear as the raw name or "custom:<name>" at runtime).
custom_providers = cfg.get("custom_providers")
if isinstance(custom_providers, list):
# Build candidate names: the provider value and the config provider
# value, both raw and with "custom:" prefix stripped/added.
candidate_names: set = set()
for p in filter(None, (provider, config_provider)):
candidate_names.add(p)
if p.startswith("custom:"):
candidate_names.add(p[len("custom:"):])
else:
candidate_names.add(f"custom:{p}")
for entry_raw in custom_providers:
if not isinstance(entry_raw, dict):
continue
entry_name = str(entry_raw.get("name") or "").strip()
if entry_name not in candidate_names:
continue
models_raw = entry_raw.get("models")
models_cfg = models_raw if isinstance(models_raw, dict) else {}
per_model_raw = models_cfg.get(model)
per_model = per_model_raw if isinstance(per_model_raw, dict) else {}
coerced = _coerce_capability_bool(per_model.get("supports_vision"))
if coerced is not None:
return coerced
return None
def _coerce_mode(raw: Any) -> str:
"""Normalize a config value into one of the valid modes."""
if not isinstance(raw, str):
return "auto"
val = raw.strip().lower()
if val in _VALID_MODES:
return val
return "auto"
def _explicit_aux_vision_override(cfg: Optional[Dict[str, Any]]) -> bool:
"""True when the user configured a specific auxiliary vision backend.
An explicit override means the user *wants* the text pipeline (they're
paying for a dedicated vision model), so we don't silently bypass it.
"""
if not isinstance(cfg, dict):
return False
aux = cfg.get("auxiliary") or {}
if not isinstance(aux, dict):
return False
vision = aux.get("vision") or {}
if not isinstance(vision, dict):
return False
provider = str(vision.get("provider") or "").strip().lower()
model = str(vision.get("model") or "").strip()
base_url = str(vision.get("base_url") or "").strip()
# "auto" / "" / blank = not explicit
if provider in {"", "auto"} and not model and not base_url:
return False
return True
def _lookup_supports_vision(
provider: str,
model: str,
cfg: Optional[Dict[str, Any]] = None,
) -> Optional[bool]:
"""Return True/False if we can resolve caps, None if unknown.
Consults the user's ``supports_vision`` override in config.yaml first
(so custom/local models declared as vision-capable don't fall through to
text routing in ``auto`` mode), then falls back to models.dev.
"""
override = _supports_vision_override(cfg, provider, model)
if override is not None:
return override
if not provider or not model:
return None
try:
from agent.models_dev import get_model_capabilities
caps = get_model_capabilities(provider, model)
except Exception as exc: # pragma: no cover - defensive
logger.debug("image_routing: caps lookup failed for %s:%s%s", provider, model, exc)
return None
if caps is None:
return None
return bool(caps.supports_vision)
def decide_image_input_mode(
provider: str,
model: str,
cfg: Optional[Dict[str, Any]],
) -> str:
"""Return ``"native"`` or ``"text"`` for the given turn.
Args:
provider: active inference provider ID (e.g. ``"anthropic"``, ``"openrouter"``).
model: active model slug as it would be sent to the provider.
cfg: loaded config.yaml dict, or None. When None, behaves as auto.
"""
mode_cfg = "auto"
if isinstance(cfg, dict):
agent_cfg = cfg.get("agent") or {}
if isinstance(agent_cfg, dict):
mode_cfg = _coerce_mode(agent_cfg.get("image_input_mode"))
if mode_cfg == "native":
return "native"
if mode_cfg == "text":
return "text"
# auto
if _explicit_aux_vision_override(cfg):
return "text"
supports = _lookup_supports_vision(provider, model, cfg)
if supports is True:
return "native"
return "text"
# Image size handling is REACTIVE rather than proactive: we attempt native
# attachment at full size regardless of provider, and rely on
# ``run_agent._try_shrink_image_parts_in_messages`` to shrink + retry if
# the provider rejects the request (e.g. Anthropic's hard 5 MB per-image
# ceiling returned as HTTP 400 "image exceeds 5 MB maximum").
#
# Why reactive: our knowledge of provider ceilings is partial and evolving
# (OpenAI accepts 49 MB+, Anthropic 5 MB, Gemini 100 MB, others unknown).
# A proactive per-provider table would be stale the moment a provider raises
# or lowers its limit, and silently degrading quality for users on providers
# that would have accepted the full image is the worse failure mode.
# The shrink-on-reject path loses 1 API call + maybe 1s of Pillow work when
# it fires, which is cheaper than permanent quality loss.
def _sniff_mime_from_bytes(raw: bytes) -> Optional[str]:
"""Detect image MIME from magic bytes. Returns None if unrecognised.
Filename-based detection (``mimetypes.guess_type``) is unreliable when
upstream platforms lie about content-type. Discord, for example, can
serve a PNG with ``content_type=image/webp`` for proxied/animated
stickers, custom emoji previews, or images uploaded via certain bots.
Anthropic strictly validates that declared media_type matches the
actual bytes and returns HTTP 400 on mismatch, so we sniff to be safe.
"""
if not raw:
return None
# PNG: 89 50 4E 47 0D 0A 1A 0A
if raw.startswith(b"\x89PNG\r\n\x1a\n"):
return "image/png"
# JPEG: FF D8 FF
if raw.startswith(b"\xff\xd8\xff"):
return "image/jpeg"
# GIF87a / GIF89a
if raw[:6] in {b"GIF87a", b"GIF89a"}:
return "image/gif"
# WEBP: "RIFF" .... "WEBP"
if len(raw) >= 12 and raw[:4] == b"RIFF" and raw[8:12] == b"WEBP":
return "image/webp"
# BMP: "BM"
if raw.startswith(b"BM"):
return "image/bmp"
# ISO-BMFF family (HEIC/HEIF/AVIF): bytes 4..8 == 'ftyp', major brand at 8..12
if len(raw) >= 12 and raw[4:8] == b"ftyp":
brand = raw[8:12]
if brand in {b"avif", b"avis"}:
return "image/avif"
if brand in {
b"heic", b"heix", b"hevc", b"hevx",
b"mif1", b"msf1", b"heim", b"heis",
}:
return "image/heic"
# TIFF: II*\0 (little-endian) or MM\0* (big-endian)
if raw[:4] in {b"II*\x00", b"MM\x00*"}:
return "image/tiff"
# ICO: 00 00 01 00 (reserved=0, type=1=icon)
if raw[:4] == b"\x00\x00\x01\x00":
return "image/x-icon"
# SVG: text-based, look for an <svg tag near the start (skip BOM/whitespace)
head = raw[:512].lstrip().lower()
if head.startswith(b"<?xml") or head.startswith(b"<svg"):
if b"<svg" in head:
return "image/svg+xml"
return None
# Formats every major vision provider (Anthropic, OpenAI, Gemini, Bedrock)
# accepts natively. Anything outside this set has to be transcoded to PNG
# before we declare media_type, otherwise the provider returns HTTP 400
# ("Could not process image" / "Unsupported image media type") and the
# whole turn fails with no salvage path.
#
# Discord (and a few other chat platforms) freely accept attachments in
# formats outside this set -- AVIF screenshots from Chromium, HEIC from
# iPhones, TIFF from scanners, BMP from old Windows tools, ICO -- so users
# do hit this in practice. SVG is vector and Pillow cannot rasterize it;
# it is skipped (logged) rather than transcoded.
_UNIVERSALLY_SUPPORTED_MIMES = frozenset({
"image/png", "image/jpeg", "image/gif", "image/webp",
})
def _transcode_to_png(raw: bytes) -> Optional[bytes]:
"""Decode arbitrary image bytes with Pillow and re-encode as PNG.
Returns None if Pillow isn't installed or can't decode the input
(rare formats, corrupted bytes, missing optional decoder plugin for
HEIC/AVIF, or vector formats like SVG). Caller falls back to skipping
the image so the rest of the turn still works.
HEIC/HEIF and AVIF need optional Pillow plugins; we try to register
them on demand and swallow ImportError so a missing plugin just
looks like 'Pillow can't decode this' rather than crashing.
"""
try:
from PIL import Image
except ImportError:
logger.info(
"image_routing: Pillow not installed; cannot transcode "
"non-standard image format to PNG. Install with `pip install Pillow` "
"(and `pillow-heif` / `pillow-avif-plugin` for those formats)."
)
return None
# Optional plugin registration. Silent on failure: an unsupported
# format will just fall through to Image.open raising below.
try:
import pillow_heif # type: ignore
pillow_heif.register_heif_opener()
except Exception:
pass
try:
import pillow_avif # type: ignore # noqa: F401 -- registers AVIF on import
except Exception:
pass
try:
from io import BytesIO
with Image.open(BytesIO(raw)) as im:
# Pick an output mode PNG can serialise. Anything other than
# the standard set gets normalised to RGBA so transparency is
# preserved where the source had it.
if im.mode not in {"RGB", "RGBA", "L", "LA", "P"}:
im = im.convert("RGBA")
buf = BytesIO()
im.save(buf, format="PNG", optimize=False)
return buf.getvalue()
except Exception as exc:
logger.info(
"image_routing: Pillow could not transcode image to PNG -- %s", exc
)
return None
def _guess_mime(path: Path, raw: Optional[bytes] = None) -> str:
"""Return image MIME type for *path*.
If *raw* bytes are provided, magic-byte sniffing wins (authoritative).
Otherwise we fall back to ``mimetypes`` then suffix-based defaults.
"""
if raw is not None:
sniffed = _sniff_mime_from_bytes(raw)
if sniffed:
return sniffed
mime, _ = mimetypes.guess_type(str(path))
if mime and mime.startswith("image/"):
return mime
# mimetypes on some Linux distros mis-maps .jpg; default to jpeg when
# the suffix looks imagey.
suffix = path.suffix.lower()
return {
".jpg": "image/jpeg",
".jpeg": "image/jpeg",
".png": "image/png",
".gif": "image/gif",
".webp": "image/webp",
".bmp": "image/bmp",
}.get(suffix, "image/jpeg")
def _file_to_data_url(path: Path) -> Optional[str]:
"""Encode a local image as a base64 data URL at its native size.
Size limits are NOT enforced here — the agent retry loop
(``run_agent._try_shrink_image_parts_in_messages``) shrinks on the
provider's first rejection. Keeping this simple means providers that
accept large images (OpenAI 49 MB+, Gemini 100 MB) don't pay a silent
quality tax just because one other provider is stricter.
Format compatibility IS handled here: if the sniffed MIME isn't one
of ``_UNIVERSALLY_SUPPORTED_MIMES`` (i.e. it's something like AVIF,
HEIC, BMP, TIFF, or ICO that some providers reject outright), we
transcode to PNG with Pillow before declaring media_type. This fixes
the user-visible "Could not process image" HTTP 400 from Anthropic on
Discord-attached AVIF/HEIC/BMP files.
Returns None if the file can't be read OR if the format isn't
universally supported AND Pillow can't transcode it (Pillow missing,
HEIC/AVIF plugin missing, vector format like SVG, corrupt bytes). The
caller reports those paths in ``skipped`` and the rest of the turn
proceeds.
"""
try:
raw = path.read_bytes()
except Exception as exc:
logger.warning("image_routing: failed to read %s%s", path, exc)
return None
mime = _guess_mime(path, raw=raw)
if mime not in _UNIVERSALLY_SUPPORTED_MIMES:
transcoded = _transcode_to_png(raw)
if transcoded is None:
logger.warning(
"image_routing: %s is %s which is not accepted by all major "
"vision providers and could not be transcoded to PNG; "
"skipping this attachment.",
path, mime,
)
return None
logger.info(
"image_routing: transcoded %s (%s) -> image/png for provider compatibility",
path.name, mime,
)
raw = transcoded
mime = "image/png"
b64 = base64.b64encode(raw).decode("ascii")
return f"data:{mime};base64,{b64}"
def build_native_content_parts(
user_text: str,
image_paths: List[str],
image_urls: Optional[List[str]] = None,
) -> Tuple[List[Dict[str, Any]], List[str]]:
"""Build an OpenAI-style ``content`` list for a user turn.
Shape:
[{"type": "text", "text": "...\\n\\n[Image attached at: /local/path]"},
{"type": "image_url", "image_url": {"url": "data:image/png;base64,..."}},
{"type": "image_url", "image_url": {"url": "https://example.com/a.png"}},
...]
Local paths are read from disk and embedded as base64 ``data:`` URLs.
Remote URLs (``http(s)://``) are passed through verbatim — the provider
fetches them server-side. The model still sees the pixels either way.
For each successfully attached image, a hint is appended to the text
part:
* local path → ``[Image attached at: <path>]``
* URL → ``[Image attached: <url>]``
The hint gives the model a string handle so MCP/skill tools that take
an image path or URL argument can be invoked on the same image without
an extra round-trip. This parallels the text-mode hint produced by
``Runner._enrich_message_with_vision`` (``vision_analyze using image_url:
<path>``) so behaviour is consistent across both image input modes.
Images are attached at their native size. If a provider rejects the
request because an image is too large (e.g. Anthropic's 5 MB per-image
ceiling), the agent's retry loop transparently shrinks and retries
once — see ``run_agent._try_shrink_image_parts_in_messages``.
Returns (content_parts, skipped). Skipped entries are local paths
that couldn't be read from disk; URLs are never skipped (they're
not validated here).
"""
skipped: List[str] = []
image_parts: List[Dict[str, Any]] = []
attached_paths: List[str] = []
attached_urls: List[str] = []
for raw_path in image_paths:
p = Path(raw_path)
if not p.exists() or not p.is_file():
skipped.append(str(raw_path))
continue
data_url = _file_to_data_url(p)
if not data_url:
skipped.append(str(raw_path))
continue
image_parts.append({
"type": "image_url",
"image_url": {"url": data_url},
})
attached_paths.append(str(raw_path))
for url in image_urls or []:
url = (url or "").strip()
if not url:
continue
image_parts.append({
"type": "image_url",
"image_url": {"url": url},
})
attached_urls.append(url)
text = (user_text or "").strip()
# If at least one image attached, build a single text part that combines
# the user's caption (or a neutral default) with one hint per image.
if attached_paths or attached_urls:
base_text = text or "What do you see in this image?"
hint_lines: List[str] = []
hint_lines.extend(f"[Image attached at: {p}]" for p in attached_paths)
hint_lines.extend(f"[Image attached: {u}]" for u in attached_urls)
combined_text = f"{base_text}\n\n" + "\n".join(hint_lines)
parts: List[Dict[str, Any]] = [{"type": "text", "text": combined_text}]
parts.extend(image_parts)
return parts, skipped
# No images successfully attached — fall back to plain text-only behaviour.
parts = []
if text:
parts.append({"type": "text", "text": text})
return parts, skipped
__all__ = [
"decide_image_input_mode",
"build_native_content_parts",
"extract_image_refs",
]