feat(agent-profile): snap vendor-locked harnesses to native model#303
Merged
Conversation
…el instead of dropping expandProfileAxes dropped (harness, model) pairs a vendor-locked harness couldn't run, so sweeping a single deepseek model silently excluded kimi-code /codex/claude-code — no head-to-head. Now a universal harness (opencode/pi/ claudish) runs every swept model; a vendor-locked harness runs the swept models in its family, or snaps to its native default (HARNESS_NATIVE_MODEL sentinel, resolved by the harness at runtime) when it supports none — so every requested harness appears in the sweep on a model it can actually run. keepIncompatible still forces every pair verbatim.
tangletools
approved these changes
Jul 2, 2026
tangletools
left a comment
tangletools
left a comment
Contributor
There was a problem hiding this comment.
✅ Auto-approved drewstone PR — e8a7550d
This PR was opened by the trusted drewstone account.
The full PR reviewer audit still runs separately and will publish findings if it detects issues.
tangletools · auto-approval · reason: drewstone_author · 2026-07-02T02:28:13Z
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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
expandProfileAxespreviously dropped harness×model cells for vendor-locked harnesses (e.g.kimi-code) that cannot run an arbitrary model. This adds a harness→model snap: such harnesses resolve to their native model via the exportedHARNESS_NATIVE_MODEL='default'sentinel instead of being silently dropped.Why
Cross-harness leaderboards (opencode vs kimi-code) need every harness to produce a runnable cell. Dropping the vendor-locked ones left the matrix incomplete. With the snap, kimi-code runs on its native Kimi model while portable harnesses run the requested model.
Change shape
Additive: new exported sentinel + new snap branch in
expandProfileAxes. No existing signature changes. Version trio bumped to 0.102.0.Verification
pnpm exec vitest run= 5320 passed / 4 skippedpnpm buildrc=0,biome checkclean,tsc --noEmitrc=0src/agent-profile.test.tsspot-check: 28 tests pass