chore(#157): remove voice-prompts feature + correct hook/skill counts#161
Merged
atlas-apex merged 3 commits intoMay 4, 2026
Merged
Conversation
Loading
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Closes #157 (sunset the voice-prompts-on-pause feature) and #77 (hook / skill count off-by-one in CHANGELOG.md + CLAUDE.md) in one bundled PR. See
AgDR-0013for the supersession rationale; AgDR-0009 stays in place as the historical record of the original Phase 1 design.main. Two months of opt-in availability produced zero adopters; this fork's local override disabled it earlier this week (Phase 1 was unreliable on the bundled macOS Daniel voice). Pre-release moment is the right time to pull cleanly rather than maintain disabled-by-default opt-in indefinitely.CHANGELOG.mdv0.3.0 stats had "17 hooks" but the actual count at v0.3.0 was 18;CLAUDE.mdhad stale "18 / 35 / 34" numbers. All updated to current reality post-removal: 24 hooks, 39 skills.Why bundled
#77's correct count depends on whether voice is in or out. Shipping #77 alone with "18" then immediately bumping to "17" when #157 lands is churn for no reason. Same shape as the previous coupled bundles — PR #152 (AgDR-0011) and PR #158 (AgDR-0012).
Closes me2resh/apexyard#157only — #77 closed manually post-merge per the single-Closes-per-PR rule.Why no adopter-facing changelog mention
The voice feature never reached
main. v1.1.0 didn't have it; v1.2.0 won't have it. From the adopter's perspective there's nothing to "retire" — they never saw the feature exist. AgDR-0013 captures the framework's internal record for future contributors. The published v1.2.0 changelog will not enumerate this removal.Testing
for t in .claude/hooks/tests/test_*.sh; do bash "$t" 2>&1 | tail -2; done→ 196/196 across 12 test files (down from 13 —test_voice_prompt_on_pause.shremoved; nothing else regressed).jq . .claude/settings.jsonandjq . .claude/project-config.defaults.jsonboth clean.Stopmatcher block was the only entry referencing voice and is now removed entirely;jq '.hooks | keys' .claude/settings.jsonreturns["PostToolUse", "PreToolUse", "SessionStart"].grep -rln "voice_prompts\|voice-prompt" .claude/ docs/ workflows/ templates/ roles/ CLAUDE.md CHANGELOG.mdreturns only AgDR-0009 (preserved historical record) and AgDR-0013 (this supersession).ls .claude/hooks/*.sh | grep -v _lib- | wc -l→ 24 ✓;ls -d .claude/skills/*/ | wc -l→ 39 ✓; both numbers now reflected in CLAUDE.md.Glossary
/setupconfig dumps. The voice feature spent two months in this state with zero adopters; this PR retires it.CLAUDE.mdand the historical v0.3.0 count inCHANGELOG.mdboth depend on the framework's hook inventory. Updating either alone risks shipping a number that's wrong by the time the next change lands. Bundling #77 with #157 closes both in one consistent set of edits.🤖 Generated with Claude Code