Repository navigation
Fix group uploads, Call agent saves, scheduled-run identity and GPT-6 model handling - #1608
Merged
Paul Lizer (paullizer) merged 9 commits intoOct 2, 2026
Conversation
Post the upload status before reading the document so the etag-guarded write in upload_to_blob no longer conflicts with its own status update (every group upload failed with Enhanced Citations on). Allow the server-bound scope_id on Call agent manifests so scoped saves pass validation. Adds regression tests and fix notes; version 0.261.217.
…p members without Graph Background runs take the owner's stored display name and email instead of the bare object ID, which also stops them overwriting the stored display name. add_user_to_group accepts a user_id with email or display name, as the REST members route does, and group conversation invites match current group members before any directory lookup. Version 0.261.218.
GPT-6 deployments reject max_tokens. Recognize gpt-6 alongside gpt-5 so response length uses max_completion_tokens and no temperature is sent. Version 0.261.219.
…restriction in the model catalog gpt-6-astra becomes a token-evidenced record: 1,050,000 context, 922,000 input and 128,000 output for OpenAI and Azure, total-generation accounting, and a Chat Completions reasoning policy of low, medium, high and xhigh. Azure Chat Completions rejects its function tools unless reasoning_effort is none, which the model rejects; toolReasoningEfforts can't express that, so it's recorded in notes. Refs #1606. Version 0.261.220.
…rt none gpt-6-sol gets verified 1,050,000 context, 922,000 input and 128,000 output limits, a reasoning policy of none through xhigh, and Chat Completions toolReasoningEfforts of none for OpenAI and Azure. OpenAI documents that its Chat Completions function calling requires none, and Azure behaved the same. Agents with tools and no effort now send none; another explicit effort fails before the request. Refs #1606. Version 0.261.221.
![github-advanced-security[bot]](https://avatars.githubusercontent.com/in/57789?s=60&v=4){kind=small}
Co-authored-by: Copilot Autofix powered by AI <62310815+github-advanced-security[bot]@users.noreply.github.com>
…call-agent Resolves the conflicts with #1607. Both branches started from 0.261.216 and used 0.261.217. VERSION stays 0.261.221, and the release notes combine both 0.261.217 entries under one heading.
…oad-call-agent' into paullizer-react-v2-fix-group-upload-call-agent
Paul Lizer (paullizer)
merged commit Oct 2, 2026
663a05c
into
paullizer-react-v2-ui
11 checks passed
Paul Lizer (paullizer)
added a commit
to paullizer/simplechat
that referenced
this pull request
Oct 2, 2026
) into 6b-1 Merged with rerere disabled. Two files conflicted: - application/single_app/config.py: kept this branch's VERSION line for the merge; the next commit renumbers 0.261.218 to 0.261.226 because V2 is now at 0.261.225. - docs/explanation/release_notes.md: this branch's section stays on top, followed by V2's 0.261.224 to 0.261.218 sections unchanged. route_backend_conversations.py merged cleanly (hunks do not overlap). Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Paul Lizer (paullizer)
added a commit
to paullizer/simplechat
that referenced
this pull request
Oct 2, 2026
V2 is now at 0.261.225 (microsoft#1612 with microsoft#1608), so this branch takes the next version. Mechanical change only: 51 occurrences in 21 files, all on lines this branch added (config.py VERSION, the release-notes heading, test headers, route-inventory docstrings and the feature docs). V2's own 0.261.218 lines from microsoft#1608 are unchanged. Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
14 of 24 tasks
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
2 participants
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.
Summary
Fixes found while running a scheduled agent workflow end to end on a SimpleChat POC deployment of the React V2 branch. Each fix has its own commit, version, fix doc and functional test.
scope_id.add_user_to_groupacceptsuser_id, and group conversation invites match current members before any Graph lookup, so scheduled runs can add and invite members.max_completion_tokensinstead ofmax_tokens, and no temperature. Refs #1605.gpt-6-astragets verified token limits and a reasoning policy (low to xhigh) in the model catalog. Its Azure Chat Completions tool restriction is recorded in notes. Refs #1606.gpt-6-soljoins the catalog with a Chat CompletionstoolReasoningEffortsofnonefor Azure and OpenAI, so agents with tools sendnoneautomatically. Refs #1606.Release notes cover all five versions.
Validation
test_group_document_blob_upload_etag.py,test_agent_delegation_scoped_action_save.py,test_workflow_background_identity.py,test_gpt6_reasoning_model_parameters.py,test_gpt6_astra_catalog_record.pyandtest_gpt6_sol_catalog_record.py. Updated:test_simplechat_agent_group_output.pyandtest_model_catalog_rebase_integration.py.test_ai_connections_capabilities.pyhas 9 failures on both, and none involves GPT-6.gpt-6-solrun with reasoning effort none.Notes
gpt-6-astrathrough Chat Completions. gpt-6-astra agents with actions fail: Azure Chat Completions rejects function tools at every reasoning effort #1606 tracks a Responses API path and an early, clear error.