Skip to content

Add git worktree management skills#8

Closed
tommy-ca wants to merge 3 commits intoEveryInc:mainfrom
tommy-ca:feature/add-git-worktree-skills
Closed

Add git worktree management skills#8
tommy-ca wants to merge 3 commits intoEveryInc:mainfrom
tommy-ca:feature/add-git-worktree-skills

Conversation

@tommy-ca
Copy link
Copy Markdown

@tommy-ca tommy-ca commented Nov 8, 2025

Summary

Extract git worktree functionality into reusable Claude Code skills to support isolated development environments and enable parallel work across the team.

Changes

New Skills (3 total)

  • git-worktree-create: Creates isolated worktrees for PRs, features, and experiments with smart defaults
  • git-worktree-manage: Manages worktree lifecycle (list, switch, cleanup, prune)
  • git-worktree-best-practices: Provides naming conventions and organizational patterns

Updated Components

  • /work command refactored to delegate to git-worktree-create skill
  • plugin.json: Version bumped to 1.1.0, added skills metadata
  • marketplace.json: Version bumped, updated descriptions
  • Created plugin README.md with comprehensive documentation

Architecture Benefits

  • ✅ Single source of truth for worktree logic (skills)
  • ✅ Commands properly delegate instead of duplicating logic
  • ✅ Zero duplication across codebase
  • ✅ Compounding engineering philosophy maintained
  • ✅ Auto-invocable skills for seamless workflow integration

Files Changed

  • Created 6 new skill files (SKILL.md + examples + scripts)
  • Updated 3 configuration/documentation files
  • Refactored 1 command file

Test Plan

  • Verified all JSON is valid
  • Confirmed file structure and permissions
  • Validated no duplication exists
  • Checked cross-references and links

Commits

  1. Add git worktree management skills
  2. Refactor /work command to use git-worktree-create skill
  3. Bump version to 1.1.0 and update documentation

tommy-ca and others added 3 commits November 8, 2025 19:57
@tommy-ca @claude
Add git worktree management skills
57ff3a7 
Introduce 3 reusable skills that encapsulate git worktree best practices:

- git-worktree-create: Create worktrees with smart defaults
  - Branch naming conventions
  - Directory structure (.worktrees/)
  - .gitignore management
  - Verification steps
  - Includes automation script and examples

- git-worktree-manage: List, remove, and maintain worktrees
  - Worktree lifecycle management
  - Cleanup and pruning
  - Status inspection

- git-worktree-best-practices: Comprehensive workflow guide
  - Common patterns and workflows
  - Team conventions
  - Troubleshooting guidance

These skills extract common patterns from commands (especially /work)
to make them reusable across agents and other commands.

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>
@tommy-ca @claude
Refactor /work command to use git-worktree-create skill
677e072 
Update the /work command to delegate worktree creation to the
git-worktree-create skill instead of duplicating the logic inline.

Changes:
- Replace inline worktree setup (43 lines) with skill invocation
- Reference skill documentation for details
- Simplify command by ~14 lines
- Maintain same functionality
- Improve maintainability (single source of truth)

The command now invokes the skill which handles:
- Branch naming
- Directory structure
- .gitignore updates
- Worktree creation
- Environment verification

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>
@tommy-ca @claude
Bump version to 1.1.0 and update documentation
6c86b24 
Release version 1.1.0 with new skills functionality.

Changes:
- Bump version: 1.0.0 → 1.1.0 (minor, additive change)
- Update plugin.json:
  - Add components.skills: 3
  - Add skills section with git-worktree skills
  - Update description to mention skills
  - Add git-worktrees keyword
- Update marketplace.json:
  - Update version to 1.1.0
  - Update description to mention skills
  - Add git-worktrees tag
- Add plugin README.md:
  - Document all components (agents, commands, skills)
  - Include usage examples
  - List all agents, commands, and skills
  - Add version history
  - Add installation instructions

This is an additive, non-breaking change that introduces reusable
skills as a new capability of the plugin.

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>
@tommy-ca
Copy link
Copy Markdown
Author

tommy-ca commented Nov 8, 2025

Closing this PR in favor of tommy-ca#1.

Reason for consolidation:

  • Reconfigured git remotes following fork-based workflow conventions
  • Fork (tommy-ca/every-marketplace) is now the canonical source for feature branches
  • New PR maintains the same commits and changes

The new PR can be found at: tommy-ca#1

🤖 Generated with Claude Code

@tommy-ca tommy-ca closed this Nov 8, 2025
armstrongl added a commit to nuggylib/compound-engineering-plugin that referenced this pull request Apr 16, 2026
@armstrongl
feat(token-improvements): brainstorm Batch 3 ideas and plan EveryInc#7 …
4beb903 
…on-demand loading

Batch 3 brainstorms (Phase 2 - Structural Dedup):
- EveryInc#7 On-Demand Loading: 5 ce-review refs (29KB) to stage-specific reads
- #3 Diff-Proportional Scaling: 4-tier reviewer cap system
- EveryInc#8 Cross-Skill Dedup: differentiate by category (keep question-tool per-file, remove native-tool dupes)
- EveryInc#21 Pipeline Semantic Dedup: critical "NEVER CODE" vs "implement" interference found

Plan EveryInc#7 complete: single-unit refactor of ce-review SKILL.md.

Also fixes 2 contract test failures from the register-mismatch pass:
- Restored behavioral ce:review reference in todo-resolve/SKILL.md
- Updated assertion to match reworded text in ce-review/SKILL.md
- Marks dead content audit (idea EveryInc#5) as execution-done
armstrongl added a commit to nuggylib/compound-engineering-plugin that referenced this pull request Apr 16, 2026
@armstrongl
feat(token-improvements): brainstorm #2 and plan #3, EveryInc#8, Ever…
e3ec793 
…yInc#21 for Batch 3

- Brainstorm #2 Lean Agent Dispatch: shared context dedup delivers
  ~140-195 KB/review by writing dispatch content to .context/ once
  instead of inlining ~16 KB per reviewer. Archetype cleanup deferred.
- Plan #3 Diff-Proportional Scaling: 5 units adding 4-tier caps
  (trivial/small/medium/large) with priority-ordered selection across
  Tier 1-4 reviewer categories.
- Plan EveryInc#8+EveryInc#21 Combined Dedup: 7 units covering AGENTS.md canonical
  sections, native tool guidance removal from 14 files, phase-scoping
  the critical "NEVER CODE" interference, semantic compression, and
  staleness checks.

Updates meta-plan tracking for all Batch 3 items.
armstrongl added a commit to nuggylib/compound-engineering-plugin that referenced this pull request Apr 16, 2026
@armstrongl
feat(token-improvements): execute Batch 3 -- plan #2, execute #3, Eve…
c355814 
…ryInc#8+EveryInc#21, #2

Batch 3 (Phase 2 - Structural Dedup) execution:

- #3 Diff-Proportional Scaling: 5 units adding cap: arg, EXECUTABLE_LINES
  counting, tier table, priority ordering, and capped team announcement to
  ce-review SKILL.md. Trivial diffs now dispatch max 8 agents (was unbounded).

- EveryInc#8+EveryInc#21 Cross-Skill Pipeline Dedup: 7 units across 25 files. AGENTS.md
  canonical Cross-Platform Interaction Convention section, native tool guidance
  removal from 14 agent/skill files, phase-scoping "NEVER CODE" interference
  (ce-plan/ce-work/ce-review), semantic compression of 3 pipeline skills,
  question-tool drift staleness check, cross-reference matrix, and
  carrying-waste manifest (97% of pipeline content is phase-specific).

- #2 Lean Agent Dispatch: planned + executed 3 units. Write-once dispatch
  context deduplicates 15.7KB of shared template/schema/scope across N
  sub-agents. ~144KB savings per 10-reviewer dispatch. Lean prompts pass
  paths instead of inlining shared content.

All 701 tests pass (4 new staleness tests). release:validate clean.
armstrongl added a commit to nuggylib/compound-engineering-plugin that referenced this pull request Apr 21, 2026
@armstrongl
feat(token-improvements): brainstorm Batch 3 ideas and plan EveryInc#7 …
24345e2 
…on-demand loading

Batch 3 brainstorms (Phase 2 - Structural Dedup):
- EveryInc#7 On-Demand Loading: 5 ce-review refs (29KB) to stage-specific reads
- #3 Diff-Proportional Scaling: 4-tier reviewer cap system
- EveryInc#8 Cross-Skill Dedup: differentiate by category (keep question-tool per-file, remove native-tool dupes)
- EveryInc#21 Pipeline Semantic Dedup: critical "NEVER CODE" vs "implement" interference found

Plan EveryInc#7 complete: single-unit refactor of ce-review SKILL.md.

Also fixes 2 contract test failures from the register-mismatch pass:
- Restored behavioral ce:review reference in todo-resolve/SKILL.md
- Updated assertion to match reworded text in ce-review/SKILL.md
- Marks dead content audit (idea EveryInc#5) as execution-done
armstrongl added a commit to nuggylib/compound-engineering-plugin that referenced this pull request Apr 21, 2026
@armstrongl
feat(token-improvements): brainstorm #2 and plan #3, EveryInc#8, Ever…
b6528f2 
…yInc#21 for Batch 3

- Brainstorm #2 Lean Agent Dispatch: shared context dedup delivers
  ~140-195 KB/review by writing dispatch content to .context/ once
  instead of inlining ~16 KB per reviewer. Archetype cleanup deferred.
- Plan #3 Diff-Proportional Scaling: 5 units adding 4-tier caps
  (trivial/small/medium/large) with priority-ordered selection across
  Tier 1-4 reviewer categories.
- Plan EveryInc#8+EveryInc#21 Combined Dedup: 7 units covering AGENTS.md canonical
  sections, native tool guidance removal from 14 files, phase-scoping
  the critical "NEVER CODE" interference, semantic compression, and
  staleness checks.

Updates meta-plan tracking for all Batch 3 items.
armstrongl added a commit to nuggylib/compound-engineering-plugin that referenced this pull request Apr 21, 2026
@armstrongl
feat(token-improvements): execute Batch 3 -- plan #2, execute #3, Eve…
c6f24cc 
…ryInc#8+EveryInc#21, #2

Batch 3 (Phase 2 - Structural Dedup) execution:

- #3 Diff-Proportional Scaling: 5 units adding cap: arg, EXECUTABLE_LINES
  counting, tier table, priority ordering, and capped team announcement to
  ce-review SKILL.md. Trivial diffs now dispatch max 8 agents (was unbounded).

- EveryInc#8+EveryInc#21 Cross-Skill Pipeline Dedup: 7 units across 25 files. AGENTS.md
  canonical Cross-Platform Interaction Convention section, native tool guidance
  removal from 14 agent/skill files, phase-scoping "NEVER CODE" interference
  (ce-plan/ce-work/ce-review), semantic compression of 3 pipeline skills,
  question-tool drift staleness check, cross-reference matrix, and
  carrying-waste manifest (97% of pipeline content is phase-specific).

- #2 Lean Agent Dispatch: planned + executed 3 units. Write-once dispatch
  context deduplicates 15.7KB of shared template/schema/scope across N
  sub-agents. ~144KB savings per 10-reviewer dispatch. Lean prompts pass
  paths instead of inlining shared content.

All 701 tests pass (4 new staleness tests). release:validate clean.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

1 participant

@tommy-ca