Handling special claude model response in github copilot provider #6369
Conversation
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
Contributor
There was a problem hiding this comment.
Pull request overview
This PR fixes a regression where tool calls stopped working for Claude models when using the GitHub Copilot provider. The issue was introduced by PR #6331 which enabled direct SSE stream consumption, but Copilot's Claude stream doesn't include function.arguments in tool calls.
Key Changes
- Removes Claude models (
claude-sonnet-4,claude-sonnet-4.5,claude-haiku-4.5) fromGITHUB_COPILOT_STREAM_MODELSso they use the completion path instead of streaming - Adds
promote_tool_choice()function to handle GitHub Copilot's behavior of returning tool_calls in non-zero index choices for Claude models - Includes unit tests covering both promotion and no-op scenarios
jamadeo
approved these changes
Jan 7, 2026
| } | ||
| return new_response; | ||
| } | ||
| } |
Collaborator
There was a problem hiding this comment.
makes sense to me, but wouldn't it be simpler to do a Vec::sort_by?
also I wonder if this would make sense to have in formats::openai::response_to_message, since that is where we select the first choice
Collaborator
Author
There was a problem hiding this comment.
@jamadeo Thanks for the review!
"sort_by" could reorder all the choices. the current approach only moves the first choice with "tool_calls" to index 0 while preserving the rest.
also I wonder if this would make sense to have in formats::openai::response_to_message, since that is where we select the first choice
Initially I implemented this logic in formats::openai::response_to_message, but later on I moved to this logic to this file to avoid complexity in formats::openai as this is the special case with github_copilot provider. When the other provider has the similar use case, we can move to formats::openai. WDYT? I can create a followup PR if it makes more sense to include this logic in formats::openai
zanesq
added a commit
that referenced
this pull request
Jan 8, 2026
* 'main' of github.com:block/goose: Fixed fonts (#6389) Update confidence levels prompt injection detection to reduce false positive rates (#6390) Add ML-based prompt injection detection (#5623) docs: update custom extensions tutorial (#6388) fix ResultsFormat error when loading old sessions (#6385) docs: add MCP Apps tutorial and documentation updates (#6384) changed z-index to make sure the search highlighter does not appear on modal overlay (#6386) Handling special claude model response in github copilot provider (#6369) fix: prevent duplicate rendering when tool returns both mcp-ui and mcp-apps resources (#6378) fix: update MCP Apps _meta.ui.resourceUri to use nested format (SEP-1865) (#6372) feat(providers): add streaming support for Google Gemini provider (#6191) Blog: edit links in mcp apps post (#6371) fix: prevent infinite loop of tool-input notifications in MCP Apps (#6374)
michaelneale
added a commit
that referenced
this pull request
Jan 8, 2026
* main: (31 commits) added validation and debug for invalid call tool result (#6368) Update MCP apps tutorial: fix _meta structure and version prereq (#6404) Fixed fonts (#6389) Update confidence levels prompt injection detection to reduce false positive rates (#6390) Add ML-based prompt injection detection (#5623) docs: update custom extensions tutorial (#6388) fix ResultsFormat error when loading old sessions (#6385) docs: add MCP Apps tutorial and documentation updates (#6384) changed z-index to make sure the search highlighter does not appear on modal overlay (#6386) Handling special claude model response in github copilot provider (#6369) fix: prevent duplicate rendering when tool returns both mcp-ui and mcp-apps resources (#6378) fix: update MCP Apps _meta.ui.resourceUri to use nested format (SEP-1865) (#6372) feat(providers): add streaming support for Google Gemini provider (#6191) Blog: edit links in mcp apps post (#6371) fix: prevent infinite loop of tool-input notifications in MCP Apps (#6374) fix: Show platform-specific keyboard shortcuts in UI (#6323) fix: we load extensions when agent starts so don't do it up front (#6350) docs: credit HumanLayer in RPI tutorial (#6365) Blog: Goose Lands MCP Apps (#6172) Claude 3.7 is out. we had some harcoded stuff (#6197) ...
wpfleger96
added a commit
that referenced
this pull request
Jan 9, 2026
* main: (89 commits) fix(google): treat signed text as regular content in streaming (#6400) Add frameDomains and baseUriDomains CSP support for MCP Apps (#6399) fix(ci): add missing dependencies to openapi-schema-check job (#6367) feat: http proxy support Add support for changing working dir and extensions in same window/session (#6057) Sort keys in canonical models (#6403) added validation and debug for invalid call tool result (#6368) Update MCP apps tutorial: fix _meta structure and version prereq (#6404) Fixed fonts (#6389) Update confidence levels prompt injection detection to reduce false positive rates (#6390) Add ML-based prompt injection detection (#5623) docs: update custom extensions tutorial (#6388) fix ResultsFormat error when loading old sessions (#6385) docs: add MCP Apps tutorial and documentation updates (#6384) changed z-index to make sure the search highlighter does not appear on modal overlay (#6386) Handling special claude model response in github copilot provider (#6369) fix: prevent duplicate rendering when tool returns both mcp-ui and mcp-apps resources (#6378) fix: update MCP Apps _meta.ui.resourceUri to use nested format (SEP-1865) (#6372) feat(providers): add streaming support for Google Gemini provider (#6191) Blog: edit links in mcp apps post (#6371) ...
wpfleger96
added a commit
that referenced
this pull request
Jan 9, 2026
* main: (89 commits) fix(google): treat signed text as regular content in streaming (#6400) Add frameDomains and baseUriDomains CSP support for MCP Apps (#6399) fix(ci): add missing dependencies to openapi-schema-check job (#6367) feat: http proxy support Add support for changing working dir and extensions in same window/session (#6057) Sort keys in canonical models (#6403) added validation and debug for invalid call tool result (#6368) Update MCP apps tutorial: fix _meta structure and version prereq (#6404) Fixed fonts (#6389) Update confidence levels prompt injection detection to reduce false positive rates (#6390) Add ML-based prompt injection detection (#5623) docs: update custom extensions tutorial (#6388) fix ResultsFormat error when loading old sessions (#6385) docs: add MCP Apps tutorial and documentation updates (#6384) changed z-index to make sure the search highlighter does not appear on modal overlay (#6386) Handling special claude model response in github copilot provider (#6369) fix: prevent duplicate rendering when tool returns both mcp-ui and mcp-apps resources (#6378) fix: update MCP Apps _meta.ui.resourceUri to use nested format (SEP-1865) (#6372) feat(providers): add streaming support for Google Gemini provider (#6191) Blog: edit links in mcp apps post (#6371) ...
fbalicchia
pushed a commit
to fbalicchia/goose
that referenced
this pull request
Jan 13, 2026
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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
Why
With the PR #6331, the tool call are no longer triggered when we using
github copilotprovider with Claude models.That change in the PR let Goose consume GitHub Copilot’s SSE stream directly. Copilot’s Claude Sonnet stream doesn’t include function.arguments, so tool calls never fire.
Why it was working before PR #6331
Before #6331 we buffered the entire stream inside the provider, so Claude effectively behaved like a non-streaming model and tools worked.
What
Type of Change
AI Assistance
Testing
Unit test and Manual testing