Repository navigation
Keep required upgrade verification beside tool selection - #470
Merged
Merged
Conversation
Place local baseline verification immediately after the tool-selection table in the shared skill and quick_start. Keep guide order, parity tests, packaging documentation and release impact aligned. Record the matched-model guardrail measurements and known scoring limitations, and distinguish current Pass 1 commands from historical references.
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Dependency-upgrade verification appears late in the shared guide and reads as an ordinary follow-up to package evidence. Move the required local checks directly after the tool-selection table in both the
githits-mcpskill andquick_start. Preserve the saved-baseline procedure and restore the explicit warning that passing existing tests does not prove compatibility.Update guide-order tests and packaging documentation, add patch release impact for both public packages, and document the matched-model guardrail comparison and its exact-string scoring limitation. Update the Pass 1 guardrail evaluation examples to use Luna consistently across comparison cohorts.
Local instruction trials remain inconclusive; this restores an earlier emphasis and does not establish reliable upgrade verification. Private upgrade fixtures and run artifacts remain outside the repository.
Validation: 5,663 Bun tests passed; build, typecheck, formatting, lint, plugin generation/checks, CLI smoke in unauthenticated mode, and MCP smoke in registration mode passed. Guide tests cover section order and the exact skill/quick_start parity. Saved guardrail reports were checked against the documented counts and marker-only failures. Local upgrade trials did not establish an improvement; no compatibility claim is made. CI passed on Ubuntu and Windows, including build/checks and Node/Bun compatibility; MCP package validation also passed. Claude review is clean after minor documentation corrections.