common/hermes
2
0
Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity

fix(copilot): mark native image requests as vision

Browse Source 
Co-authored-by: dhabibi <9087935+dhabibi@users.noreply.github.com>
This commit is contained in:
hermes-agent-dhabibi
2026-04-27 13:48:51 +00:00
committed by Teknium
parent 8402ba150e
commit aa53fb661a
2 changed files with 144 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

52
run_agent.py
View File

@@ -5258,7 +5258,39 @@ class AIAgent:
logger.debug("Dead connection check error: %s", exc)
return False
def _create_request_openai_client(self, *, reason: str) -> Any:
@staticmethod
def _api_kwargs_have_image_parts(api_kwargs: dict) -> bool:
"""Return True when the outbound request still contains native image parts."""
if not isinstance(api_kwargs, dict):
return False
candidates = []
messages = api_kwargs.get("messages")
if isinstance(messages, list):
candidates.extend(messages)
# Responses API payloads use `input`; after conversion, image parts can
# still be present there instead of in `messages`.
response_input = api_kwargs.get("input")
if isinstance(response_input, list):
candidates.extend(response_input)
def _contains_image(value: Any) -> bool:
if isinstance(value, dict):
ptype = value.get("type")
if ptype in {"image_url", "input_image"}:
return True
return any(_contains_image(v) for v in value.values())
if isinstance(value, list):
return any(_contains_image(v) for v in value)
return False
return any(_contains_image(item) for item in candidates)
def _copilot_headers_for_request(self, *, is_vision: bool) -> dict:
from hermes_cli.copilot_auth import copilot_request_headers
return copilot_request_headers(is_agent_turn=True, is_vision=is_vision)
def _create_request_openai_client(self, *, reason: str, api_kwargs: Optional[dict] = None) -> Any:
from unittest.mock import Mock
primary_client = self._ensure_primary_openai_client(reason=reason)
@@ -5266,6 +5298,11 @@ class AIAgent:
return primary_client
with self._openai_client_lock():
request_kwargs = dict(self._client_kwargs)
if (
base_url_host_matches(str(request_kwargs.get("base_url", "")), "api.githubcopilot.com")
and self._api_kwargs_have_image_parts(api_kwargs or {})
):
request_kwargs["default_headers"] = self._copilot_headers_for_request(is_vision=True)
return self._create_openai_client(request_kwargs, reason=reason, shared=False)
def _close_request_openai_client(self, client: Any, *, reason: str) -> None:
@@ -5808,7 +5845,10 @@ class AIAgent:
def _call():
try:
if self.api_mode == "codex_responses":
request_client_holder["client"] = self._create_request_openai_client(reason="codex_stream_request")
request_client_holder["client"] = self._create_request_openai_client(
reason="codex_stream_request",
api_kwargs=api_kwargs,
)
result["response"] = self._run_codex_stream(
api_kwargs,
client=request_client_holder["client"],
@@ -5840,7 +5880,10 @@ class AIAgent:
raise
result["response"] = normalize_converse_response(raw_response)
else:
request_client_holder["client"] = self._create_request_openai_client(reason="chat_completion_request")
request_client_holder["client"] = self._create_request_openai_client(
reason="chat_completion_request",
api_kwargs=api_kwargs,
)
result["response"] = request_client_holder["client"].chat.completions.create(**api_kwargs)
except Exception as e:
result["error"] = e
@@ -6183,7 +6226,8 @@ class AIAgent:
),
}
request_client_holder["client"] = self._create_request_openai_client(
reason="chat_completion_stream_request"
reason="chat_completion_stream_request",
api_kwargs=stream_kwargs,
)
# Reset stale-stream timer so the detector measures from this
# attempt's start, not a previous attempt's last chunk.