Fix instruction template modality rendering#2030
Merged
Merged
Conversation
🦋 Changeset detectedLatest commit: 3ff9643 The changes in this PR will be included in the next version bump. This PR includes changesets to release 36 packages
Not sure what this means? Click here to learn what changesets are. Click here if you're a maintainer who wants to add another changeset to this PR |
toubatbrian
approved these changes
Jul 14, 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Instructions.tplaudio/text variants to a single audio-only representation.renderInstructions(..., modality)does not duplicate template output.Source diff coverage
Source diff coverage
livekit-agents/livekit/agents/llm/chat_context.pyagents/src/llm/chat_context.tsresolve_templatebehavior to the JS counterpart,Instructions.tpl. JS stores fully renderedaudio/textvariants plus a currentvalueinstead of Python'scommonplus modality suffixes, so the target change collapses identical rendered variants by omittingtextwhenaudio === text.tests/test_chat_ctx.pyagents/src/llm/chat_context.test.tsInstructions.tplandrenderInstructions, without adding unrelated new coverage.Verification
pnpm test -- agents/src/llm/chat_context.test.tspasses.pnpm buildpasses.pnpm lintpasses, with existing warnings.pnpm test -- agentswas run; it fails in existing AMD tests underagents/src/voice/amd.test.ts.pnpm test -- agents/src/voice/amd.test.tsreproduces the same AMD failures in isolation, confirming they are unrelated to this instruction-template port.No source infrastructure gap was identified.
Ported from livekit/agents#6414
Original PR description
Fixes #6407
Cue voice E2E
sid_7249cd3cd100at head3ff96430.PR2030_VERIFIED common=1 audio=1 text=0, proving one common marker, one audio marker, and no text marker.~/.cue-cli/sessions/sid_7249cd3cd100/.