fix: backfill empty codex output in auxiliary client#5730
Merged
Conversation
The _CodexCompletionsAdapter (used for compression, vision, web_extract, session_search, and memory flush when on the codex provider) streamed responses but discarded all events with 'for _event in stream: pass'. When get_final_response() returned empty output (the same chatgpt.com backend-api shape change), auxiliary calls silently returned None content. Now collects response.output_item.done and text deltas during streaming and backfills empty output — same pattern as _run_codex_stream(). Tested live against chatgpt.com/backend-api/codex with OAuth.
Tommyeds
pushed a commit
to Tommyeds/hermes-agent
that referenced
this pull request
Apr 12, 2026
The _CodexCompletionsAdapter (used for compression, vision, web_extract, session_search, and memory flush when on the codex provider) streamed responses but discarded all events with 'for _event in stream: pass'. When get_final_response() returned empty output (the same chatgpt.com backend-api shape change), auxiliary calls silently returned None content. Now collects response.output_item.done and text deltas during streaming and backfills empty output — same pattern as _run_codex_stream(). Tested live against chatgpt.com/backend-api/codex with OAuth.
3 tasks
angelburgosrosado
pushed a commit
to angelburgosrosado/hermes-agent
that referenced
this pull request
Apr 27, 2026
The _CodexCompletionsAdapter (used for compression, vision, web_extract, session_search, and memory flush when on the codex provider) streamed responses but discarded all events with 'for _event in stream: pass'. When get_final_response() returned empty output (the same chatgpt.com backend-api shape change), auxiliary calls silently returned None content. Now collects response.output_item.done and text deltas during streaming and backfills empty output — same pattern as _run_codex_stream(). Tested live against chatgpt.com/backend-api/codex with OAuth.
02356abc
pushed a commit
to 02356abc/hermes-agent
that referenced
this pull request
May 14, 2026
The _CodexCompletionsAdapter (used for compression, vision, web_extract, session_search, and memory flush when on the codex provider) streamed responses but discarded all events with 'for _event in stream: pass'. When get_final_response() returned empty output (the same chatgpt.com backend-api shape change), auxiliary calls silently returned None content. Now collects response.output_item.done and text deltas during streaming and backfills empty output — same pattern as _run_codex_stream(). Tested live against chatgpt.com/backend-api/codex with OAuth.
olympus-terminal
pushed a commit
to olympus-terminal/hermes-agent
that referenced
this pull request
May 16, 2026
The _CodexCompletionsAdapter (used for compression, vision, web_extract, session_search, and memory flush when on the codex provider) streamed responses but discarded all events with 'for _event in stream: pass'. When get_final_response() returned empty output (the same chatgpt.com backend-api shape change), auxiliary calls silently returned None content. Now collects response.output_item.done and text deltas during streaming and backfills empty output — same pattern as _run_codex_stream(). Tested live against chatgpt.com/backend-api/codex with OAuth.
gweeteve
pushed a commit
to gweeteve/hermes-agent
that referenced
this pull request
Jun 2, 2026
The _CodexCompletionsAdapter (used for compression, vision, web_extract, session_search, and memory flush when on the codex provider) streamed responses but discarded all events with 'for _event in stream: pass'. When get_final_response() returned empty output (the same chatgpt.com backend-api shape change), auxiliary calls silently returned None content. Now collects response.output_item.done and text deltas during streaming and backfills empty output — same pattern as _run_codex_stream(). Tested live against chatgpt.com/backend-api/codex with OAuth.
Egavasyug
pushed a commit
to Egavasyug/hermes-agent
that referenced
this pull request
Jun 10, 2026
The _CodexCompletionsAdapter (used for compression, vision, web_extract, session_search, and memory flush when on the codex provider) streamed responses but discarded all events with 'for _event in stream: pass'. When get_final_response() returned empty output (the same chatgpt.com backend-api shape change), auxiliary calls silently returned None content. Now collects response.output_item.done and text deltas during streaming and backfills empty output — same pattern as _run_codex_stream(). Tested live against chatgpt.com/backend-api/codex with OAuth.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The auxiliary client's
_CodexCompletionsAdapter(used for compression, vision, web_extract summarization, session_search, memory flush — everything that goes through the auxiliary LLM path when on the codex provider) had the same empty-output bug as the main agent loop.It streamed responses but discarded all events with
for _event in stream: pass, then calledget_final_response()which returnsoutput=[]from the chatgpt.com backend. Auxiliary calls silently returnedcontent=None.Now collects
response.output_item.doneand text deltas during streaming and backfills empty output — same pattern as_run_codex_stream().Tested live against chatgpt.com/backend-api/codex with OAuth — log confirms:
Codex auxiliary: backfilled 1 output items from stream events