feat: allow experimental task model selection - #13557
Merged
marius-kilocode merged 2 commits intoAug 31, 2026
Merged
Conversation
Contributor
Code Review SummaryStatus: No Issues Found | Recommendation: Merge Files Reviewed (39 files)
Reviewed by grok-4.6 · Input: 466.8K · Output: 37K · Cached: 1.9M Review guidance: REVIEW.md from base branch |
marius-kilocode
enabled auto-merge
August 28, 2026 17:14
WebReflection
approved these changes
Aug 31, 2026
marius-kilocode
deleted the
enable-configurable-subagent-model-selection
branch
August 31, 2026 08:00
This was referenced Aug 31, 2026
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What Problem This Solves
Internal Task subagents cannot select a model, provider, or reasoning effort for an individual task. An orchestrator must use the configured subagent defaults or switch to Agent Manager sessions, even when an internal subagent is the correct workflow.
Why This Change Was Made
Add the opt-in
experimental.task_model_selectionconfig flag and expose it in Kilo's Experimental settings. It uses the existing Save/Discard and backend config refresh path, rather than a startup-only environment variable. Saving the setting refreshes cached Task schemas without restarting the backend.Task and Agent Manager share model lookup, provider preference, ambiguity handling, and variant validation. Invalid explicit selections fail before creating a child session instead of silently falling back to another model.
User Impact
model,provider, andvariantoverrides. An orchestrator can useagent_manager_modelsto discover valid choices.Evidence
low,medium, andhighreasoning, and the saved child session metadata matched. This verifies local routing and serialization, not whether a live upstream provider honors a reasoning setting.