[genai-openai] Responses: reasoning items passed back as input are dropped
Maintainers usually reply within 1 day
Assessment
- Difficulty
- 2/5
- Estimated time
- 1-3 hours
- Newbie friendliness
- 74/100
Research direction
Locate _get_input_message and the existing _extract_reasoning_parts helper in the Responses instrumentation. Trace how tool calls, tool outputs, and role messages become gen_ai.input.messages, then verify the reasoning-input path alongside the existing tests. Done means prior reasoning items appear as assistant messages with their reasoning parts in the instrumentation output.
Written by the indexing model from the issue text.
Description
[!NOTE]
This issue was created with AI assistance and needs to be verified.
On multi-turn Responses calls, apps feed prior output, including type: "reasoning" items, back into responses.create(input=[...]). _get_input_message handles tool calls, tool outputs, and role messages only, so reasoning input items are skipped and missing from gen_ai.input.messages.
Fix: map reasoning input items to an assistant message with ReasoningPart, reusing _extract_reasoning_parts.
- Dominant language
- Python
- Stars
- 42
- Forks
- 74
- Avg merge
- 2d 2h
- Merged PRs (30d)
- 126
Getting set up
- No Dockerfile or Docker Compose file
- Has a pull request template
- Read the contributing guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
More from open-telemetry/opentelemetry-python-genai
-
documentation
Difficulty 2/5 1-3 hours Newbie friendliness 76/100
open-telemetry/opentelemetry-python-genai#857 · 1 comment · 1 reaction ·
Maintainers usually reply within 1 day
-
[genai-bedrock] converse with messages passed as a tuple: gen_ai.input.messages is not recordedOpen
Difficulty 2/5 1-3 hours Newbie friendliness 76/100
open-telemetry/opentelemetry-python-genai#849 ·
Maintainers usually reply within 1 day
-
Difficulty 2/5 1-3 hours Newbie friendliness 82/100
open-telemetry/opentelemetry-python-genai#848 ·
Maintainers usually reply within 1 day
-
[google-genai] uninstrument() raises AssertionError because __init__ resets the method snapshot on the singleton instrumentor; calls stay tracedPossibly taken @VimalN2005 claimed this 3 days ago. Open
Difficulty 2/5 1-3 hours Newbie friendliness 74/100
open-telemetry/opentelemetry-python-genai#845 ·
Maintainers usually reply within 1 day
-
[genai-openai] Responses: record computer_call server toolPossibly taken @1fanwang claimed this 6 days ago. Openenhancement
Difficulty 1/5 Under an hour Newbie friendliness 76/100
open-telemetry/opentelemetry-python-genai#831 ·
Maintainers usually reply within 1 day
All issues in open-telemetry/opentelemetry-python-genai
Similar issues
-
HTML: <template> content is extracted as document textPossibly taken @ryanmeowy claimed this today. Openbug html
Difficulty 1/5 Under an hour Newbie friendliness 82/100
docling-project/docling#4714 · 2 comments ·
Maintainers usually reply within 1 day
-
[BUG] Qdrant RAG client applies score_threshold to raw cosine similarity, not the 0-1 score it returnsPossibly taken @roydonsequeira claimed this today. Openbug
Difficulty 2/5 1-3 hours Newbie friendliness 70/100
Maintainers usually reply within 1 day
-
Difficulty 2/5 1-3 hours Newbie friendliness 72/100
ashhart/TensorFold#536 ·
Maintainers usually reply within 1 day
-
area/install-update comp/cli duplicate P2 python:uv sweeper:risk-compatibility type/bug
Difficulty 1/5 Under an hour Newbie friendliness 62/100
NousResearch/hermes-agent#135440 · 1 comment ·
Maintainers usually reply within 1 day
-
bug
Difficulty 2/5 1-3 hours Newbie friendliness 85/100
Deepak3699/Ai_Mentor#244 ·
Maintainers usually reply within 1 day