- Fixed condition to properly detect IRSA environments using AWS_WEB_IDENTITY_TOKEN_FILE
- Skip role assumption when already running as target role in IRSA environment
- Prevents unnecessary AWS API calls that cause 'root account cannot assume role' errors
- Resolves failing test test_auth_with_aws_role_same_role_irsa
The original issue was that aws_access_key_id and aws_secret_access_key were being
populated from environment variables even when passed as None, causing the IRSA
detection condition to fail. The fix checks for IRSA-specific environment variables
instead of relying on the absence of explicit credentials.
Fixes#13417
* feat(organization_endpoints.py): expose new `/organization/delete` endpoint. Cascade org deletion to member, teams and keys
Ensures any org deletion is handled correctly
* test(test_organizations.py): add simple test to ensure org deletion works
* feat(organization_endpoints.py): expose /organization/update endpoint, and define response models for org delete + update
* fix(organizations.tsx): support org delete on UI + move org/delete endpoint to use DELETE
* feat(organization_endpoints.py): support `/organization/member_update` endpoint
Allow admin to update member's role within org
* feat(organization_endpoints.py): support deleting member from org
* test(test_organizations.py): add e2e test to ensure org member flow works
* fix(organization_endpoints.py): fix code qa check
* fix(schema.prisma): don't introduce ondelete:cascade - breaking change
* docs(organization_endpoints.py): document missing params
- Fix supports_reasoning() call to use lowercase model names for proper lookup
- Remove custom_llm_provider parameter as model registry entries are provider-agnostic
- Update tests to use full model names with date stamps (required for supports_reasoning)
- Add test coverage for models without extended thinking support
The provider_config_manager already handles github_copilot provider
through LlmProviders.GITHUB_COPILOT mapping, making the explicit
check unnecessary.
Only models in the 4 family and 3-7 family support extended thinking features.
Previously all models would incorrectly receive these parameters.
Now uses supports_reasoning() to check model registry for actual capability.
Responses API - add default api version for openai responses api calls + Openrouter - fix claude-sonnet-4 on openrouter + Azure - Handle `openai/v1/responses`
* fix(access group): allow access group on mcp tool retrieval
* fix(test): fix broken tests and add test case for access group
* fix(mypy): fix typing issues
* fix(memory file): add content type to in memory file
- Add dynamic parameter support for anthropic models through GitHub Copilot
- Include thinking parameter for anthropic model compatibility
- Support reasoning_effort parameter for both anthropic and reasoning models
- Update test coverage for parameter validation logic
- Ensure proper parameter filtering based on model type
- Add github_copilot case to get_supported_openai_params function
- Implement get_supported_openai_params method in GithubCopilotConfig
- Dynamically add thinking and reasoning_effort params for Anthropic models
- Add comprehensive tests for parameter support validation
- Ensure case-insensitive model detection for parameter inclusion
Fixes UnsupportedParamsError when using advanced reasoning parameters
with Anthropic models through GitHub Copilot proxy.
