Repository navigation
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Removing an inactive custom provider keeps the current provider and model. Removing the active one restores the signed-in Autohand account, or asks for a replacement; cancelling keeps the active provider intact.
OpenAI and Azure now advertise a 128-tool limit, as does OpenRouter for openai/ models. The existing runtime selection limit bounds those requests. Azure reasoning models use the shared completion-token predicate and omit temperature; gpt-4o keeps its existing request parameters.
Covers F-0407, F-0409 and the initial F-0419 parameter fix. Related: #719 and #720; close those only after the fix is released.
Validation: 203 focused provider/agent tests pass, the local build passes, and the new real-terminal provider-removal journey passes. Lint and typecheck passed in the required proof run. Full proof is not green: 13,769 passed / 22 failed before environment corrections; the affected failure-file rerun is 130 passed / 5 failed. All four memory/launcher failures reproduce on unchanged released 0.9.10 across the baseline checks. The remaining peer-capacity timeout passes on both trees when isolated. The obsolete exact-capability assertion is corrected and passes.
Draft remains unready pending aggregate validation. The Azure Responses transport follow-up and a real Azure tool turn are not complete; no test endpoint or API key is configured locally. No CLI release, version bump, merge or publication is included.