feat: Conductor 0.2.0 - Core Extraction & Multi-Platform Integration #37
Conversation
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
Introduces Claude Code plugin manifests, slash command definitions, and Agent Skills-compatible skill files for Conductor. Adds detailed documentation for Claude Code usage and interoperability with Gemini CLI and Agent Skills. Updates README to reflect multi-platform support and installation options.
This adds Conductor as a reusable AI agent skill following the skills standard adopted by Claude CLI, OpenCode, and Codex. The skill provides: - SKILL.md with core concepts and quick start guide - Protocol references for each command (setup, newTrack, implement, status, revert) - Install script that copies assets from templates/ Benefits: - Global installation (~/.claude/skills/conductor/) - Works with multiple AI CLIs (Claude, OpenCode, Codex) - Auto-discovery by AI agents when relevant - Reuses existing templates (no duplication)
- Make description more action-oriented (create feature, write spec, etc.) - Add setup mention for unconfigured projects - Installer now asks user where to install (OpenCode/Claude CLI/Both)
- Remove skill/references/*.md (790 lines of duplication) - SKILL.md now points directly to commands/conductor/*.toml - install.sh creates symlinks to repo's commands/ and templates/ - Update README with interactive installer options This ensures zero duplication - the skill references the original TOML files and templates, and updates automatically with git pull.
The original TOML protocols reference Gemini CLI commands like /conductor:implement. Added rule to instruct the agent to use agnostic language instead (e.g., 'ask me to implement').
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 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.
Conductor 0.2.0: Core Extraction & Multi-Platform Integration
Summary:
This major update completes the transition of Conductor into a modular monorepo architecture. The core logic has been extracted into a standalone Python package (
conductor-core), and all platform adapters (Gemini CLI, VS Code, Claude Code, Qwen Code, Codex) have been refactored or implemented to use this centralized logic.Key Changes:
conductor-core):git notesfor automated, auditable task summaries and phase verification reports.conductor-gemini):conductor-core.implement,status,archive, andrevertcommands.setup,status,new-track, andimplementcommands.qwen-extension.jsonand verified remote installation via--ref.install.sh) to support Codex, OpenCode, and Claude CLI..github/workflows/ci.yml) providing full test coverage on every push.npm cifor reliable builds.npm ci.0.2.0.CHANGELOG.md.Testing/Verification Performed:
conductor-coreandconductor-geminiwith 100% pass rate.scripts/smoke_test.pyto verify the fullCLI -> Core -> Gitlifecycle.scripts/validate_platforms.py --sync.feature/foundation-corebranch.The project is now in a highly robust state, with all adapters in perfect sync and the core logic thoroughly tested. All CI checks are passing.