Skip to content

docs(auto-model): clarify KiloClaw plan-mode routing#1666

Closed
kilo-code-bot[bot] wants to merge 1 commit intomainfrom
docs/clarify-kiloclaw-plan-model-routing
Closed

docs(auto-model): clarify KiloClaw plan-mode routing#1666
kilo-code-bot[bot] wants to merge 1 commit intomainfrom
docs/clarify-kiloclaw-plan-model-routing

Conversation

@kilo-code-bot
Copy link
Copy Markdown
Contributor

@kilo-code-bot kilo-code-bot bot commented Mar 27, 2026

Summary

  • Added clarifying comments to src/lib/kilo-auto-model.ts explaining how KiloClaw requests are routed through the auto-model system. KiloClaw forces the mode to 'plan' (overriding the client-sent mode header) to ensure it always resolves to the highest-capability reasoning model: Claude Opus for frontier, Kimi for balanced, both with reasoning enabled.
  • Extracted the inline ternary into a named effectiveMode variable for readability, and added cross-reference comments on the plan entries in both FRONTIER_MODE_TO_MODEL and BALANCED_MODE_TO_MODEL routing tables so readers understand these entries serve KiloClaw in addition to the regular plan editor mode.

Verification

  • Verified the diff is comment-only and a local variable extraction with no behavioral change.
  • Typecheck could not run due to missing node_modules/tsgo in the environment, but the change is purely additive comments and a trivial refactor (extracting an inline expression to a named const).

Visual Changes

N/A

Reviewer Notes

The key insight documented here: applyResolvedAutoModel overrides the mode to 'plan' when featureHeader === 'kiloclaw' because KiloClaw is an autonomous cloud agent that doesn't use interactive editor modes. This was previously undocumented and non-obvious to anyone reading the routing tables or the resolve logic.

);
// KiloClaw is an autonomous cloud agent without user-selected editor modes, so we
// force mode to 'plan' to route it to the highest-capability reasoning model
// (Opus for frontier, Kimi for balanced) with reasoning enabled and high verbosity.
Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

SUGGESTION: The comment overstates the balanced plan route

BALANCED_MODE_TO_MODEL.plan enables reasoning, but it does not set verbosity, so "with reasoning enabled and high verbosity" reads as if both the frontier and balanced KiloClaw paths get high verbosity. Rephrasing this to scope high verbosity to the frontier route would keep the documentation aligned with the implementation.

@kilo-code-bot
Copy link
Copy Markdown
Contributor Author

kilo-code-bot bot commented Mar 27, 2026

Code Review Summary

Status: 1 Issue Found | Recommendation: Address before merge

Overview

Severity Count
CRITICAL 0
WARNING 0
SUGGESTION 1
Issue Details (click to expand)

SUGGESTION

File Line Issue
src/lib/kilo-auto-model.ts 262 Comment says KiloClaw gets high verbosity, but the balanced plan route does not set verbosity.
Other Observations (not in diff)

Issues found in unchanged code that cannot receive inline comments:

N/A

Files Reviewed (1 files)
  • src/lib/kilo-auto-model.ts - 1 issue

Fix these issues in Kilo Cloud


Reviewed by gpt-5.4-20260305 · 155,341 tokens

@chrarnoldus chrarnoldus deleted the docs/clarify-kiloclaw-plan-model-routing branch March 27, 2026 18:42
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant