[PR #23] [CLOSED] feat: Ralph Orchestrator v2.0 - Complete Self-Improvement & Orchestration Architecture #97

New issue

Closed

opened 2026-02-27 10:22:13 +03:00 by kerem · 0 comments

kerem commented

2026-02-27 10:22:13 +03:00

Owner

📋 Pull Request Information

Original PR: https://github.com/mikeyobrien/ralph-orchestrator/pull/23
Author: @krzemienski
Created: 1/4/2026
Status: ❌ Closed

Base: main ← Head: feat/orchestration

📝 Commits (10+)

461d7d4 feat: Self-improvement runner + User-Collaborative Validation Gates
c36f96c feat(tui): complete TUI implementation with 149 tests, 87% coverage
744a1b9 chore: cleanup artifacts and add env example
93ea869 docs: update README with TUI feature and test counts
6b7086a fix(tests): correct test mocking for ACP adapter availability
4a21192 feat(config): add enable_validation to RalphConfig with validators
bbb8bcf feat(onboarding): implement SettingsLoader for Claude Code settings
6308c8d feat(onboarding): implement ProjectScanner for project discovery
bfa732a feat(onboarding): Add AgentAnalyzer for Claude-powered project analysis
fe5894d Ralph checkpoint 3

📊 Changes

219 files changed (+86753 additions, -1518 deletions)

View changed files

➖ .agent/scratchpad.md (+0 -12)
➕ .env.example (+76 -0)
📝 .gitignore (+17 -1)
➕ .playwright-mcp/ralph-web-ui-no-auth.png (+0 -0)
➕ .playwright-mcp/ralph-web-ui-working.png (+0 -0)
➖ PROMPT.md (+0 -914)
📝 README.md (+73 -11)
➕ docs/TODO-external-state-management.md (+75 -0)
➕ docs/codebase-summary.md (+491 -0)
➕ docs/designs/2026-01-04-agent-harness.md (+286 -0)
➕ docs/designs/2026-01-04-onboarding-architecture.md (+404 -0)
➕ docs/future-features.md (+21 -0)
➕ docs/guide/onboarding.md (+461 -0)
➕ docs/guide/validation.md (+298 -0)
➕ docs/plans/260104-subagent-orchestration-architecture.md (+669 -0)
➕ docs/project-overview-pdr.md (+289 -0)
➕ examples/orchestration.yml (+67 -0)
➕ examples/self-improvement.yml (+68 -0)
➕ plans/reports/debugger-260104-0956-iteration-loop-analysis.md (+205 -0)
➕ plans/reports/debugger-260104-0956-metrics-tracking-gap.md (+258 -0)

...and 80 more files

📄 Description

Summary

This PR consolidates 95 commits of work on Ralph Orchestrator v2.0, including:

Self-Improvement Project (7 Phases) ✅ ALL VALIDATED

Phase	Feature	Status
00	TUI Verification	✅ Validated
01	Process Isolation	✅ Validated
02	Daemon Mode	✅ Validated
03	REST API	✅ Validated
04	Mobile Foundation	✅ Validated
05	Mobile Dashboard	✅ Validated
06	Mobile Control	✅ Validated

Orchestration Architecture (O0-O5) ✅ ALL COMPLETE

O0: Run Isolation & State Management (RunManager)
O1: Subagent Types & Profiles (4 specialized profiles)
O2: Skill Discovery (scans ~/.claude/skills/)
O3: MCP Tool Discovery (reads ~/.mcp.json)
O4: Coordination Protocol (file-based subagent communication)
O5: Integration & Subagent Spawning (OrchestrationManager)

New Modules

src/ralph_orchestrator/orchestration/ - 5 modules for subagent coordination
src/ralph_orchestrator/onboarding/ - 6 modules for project onboarding
src/ralph_orchestrator/daemon/ - Background daemon mode
src/ralph_orchestrator/run_manager.py - Run isolation with unique IDs

Mobile App (ralph-mobile/)

Expo React Native app for iOS/Android
API layer: 100% complete with all endpoints
Helper layer: 100% complete with validation
UI layer: ~30% complete (Dashboard works, other screens stubbed)
Tests: 13 test files (mocked)

Test Coverage

1696 total tests (up from ~250 baseline)
105 orchestration-specific tests
~350 onboarding tests
All passing

Documentation

Architecture design docs
Validation evidence for all phases
Comprehensive acceptance criteria

Test plan

All 1696 tests pass (uv run pytest)
TUI launches correctly
Daemon mode works (start/stop/status)
REST API endpoints respond
Orchestration module imports correctly
Mobile app builds (Expo) - manual verification

_{🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.}

## 📋 Pull Request Information **Original PR:** https://github.com/mikeyobrien/ralph-orchestrator/pull/23 **Author:** [@krzemienski](https://github.com/krzemienski) **Created:** 1/4/2026 **Status:** ❌ Closed **Base:** `main` ← **Head:** `feat/orchestration` --- ### 📝 Commits (10+) - [`461d7d4`](https://github.com/mikeyobrien/ralph-orchestrator/commit/461d7d46c25d4eda4af13f9f22227ee3624707d3) feat: Self-improvement runner + User-Collaborative Validation Gates - [`c36f96c`](https://github.com/mikeyobrien/ralph-orchestrator/commit/c36f96cdc021a5456c8f2059bf9b363cf4822969) feat(tui): complete TUI implementation with 149 tests, 87% coverage - [`744a1b9`](https://github.com/mikeyobrien/ralph-orchestrator/commit/744a1b9769f5326d90e09f1a1a67fdf3e27b30af) chore: cleanup artifacts and add env example - [`93ea869`](https://github.com/mikeyobrien/ralph-orchestrator/commit/93ea8691dcbd7d5ca6a9c393a13779c82c258eb2) docs: update README with TUI feature and test counts - [`6b7086a`](https://github.com/mikeyobrien/ralph-orchestrator/commit/6b7086a87e798ab3b1f1ef3fcc6eeaabdeb43a86) fix(tests): correct test mocking for ACP adapter availability - [`4a21192`](https://github.com/mikeyobrien/ralph-orchestrator/commit/4a21192c4018d65745c68ec182b4a7b05ddd264a) feat(config): add enable_validation to RalphConfig with validators - [`bbb8bcf`](https://github.com/mikeyobrien/ralph-orchestrator/commit/bbb8bcf7701c75d6862a99016dbaad25fbc0c776) feat(onboarding): implement SettingsLoader for Claude Code settings - [`6308c8d`](https://github.com/mikeyobrien/ralph-orchestrator/commit/6308c8d90d46a11bd35960bdf0cc190de37c7853) feat(onboarding): implement ProjectScanner for project discovery - [`bfa732a`](https://github.com/mikeyobrien/ralph-orchestrator/commit/bfa732aecc05fc9f88c6e85eadd949fdedf1c567) feat(onboarding): Add AgentAnalyzer for Claude-powered project analysis - [`fe5894d`](https://github.com/mikeyobrien/ralph-orchestrator/commit/fe5894d47a6ee1ba21fefd50e2bea9f0b90fe971) Ralph checkpoint 3 ### 📊 Changes **219 files changed** (+86753 additions, -1518 deletions) <details> <summary>View changed files</summary> ➖ `.agent/scratchpad.md` (+0 -12) ➕ `.env.example` (+76 -0) 📝 `.gitignore` (+17 -1) ➕ `.playwright-mcp/ralph-web-ui-no-auth.png` (+0 -0) ➕ `.playwright-mcp/ralph-web-ui-working.png` (+0 -0) ➖ `PROMPT.md` (+0 -914) 📝 `README.md` (+73 -11) ➕ `docs/TODO-external-state-management.md` (+75 -0) ➕ `docs/codebase-summary.md` (+491 -0) ➕ `docs/designs/2026-01-04-agent-harness.md` (+286 -0) ➕ `docs/designs/2026-01-04-onboarding-architecture.md` (+404 -0) ➕ `docs/future-features.md` (+21 -0) ➕ `docs/guide/onboarding.md` (+461 -0) ➕ `docs/guide/validation.md` (+298 -0) ➕ `docs/plans/260104-subagent-orchestration-architecture.md` (+669 -0) ➕ `docs/project-overview-pdr.md` (+289 -0) ➕ `examples/orchestration.yml` (+67 -0) ➕ `examples/self-improvement.yml` (+68 -0) ➕ `plans/reports/debugger-260104-0956-iteration-loop-analysis.md` (+205 -0) ➕ `plans/reports/debugger-260104-0956-metrics-tracking-gap.md` (+258 -0) _...and 80 more files_ </details> ### 📄 Description ## Summary This PR consolidates **95 commits** of work on Ralph Orchestrator v2.0, including: ### Self-Improvement Project (7 Phases) ✅ ALL VALIDATED | Phase | Feature | Status | |-------|---------|--------| | 00 | TUI Verification | ✅ Validated | | 01 | Process Isolation | ✅ Validated | | 02 | Daemon Mode | ✅ Validated | | 03 | REST API | ✅ Validated | | 04 | Mobile Foundation | ✅ Validated | | 05 | Mobile Dashboard | ✅ Validated | | 06 | Mobile Control | ✅ Validated | ### Orchestration Architecture (O0-O5) ✅ ALL COMPLETE - **O0**: Run Isolation & State Management (RunManager) - **O1**: Subagent Types & Profiles (4 specialized profiles) - **O2**: Skill Discovery (scans ~/.claude/skills/) - **O3**: MCP Tool Discovery (reads ~/.mcp.json) - **O4**: Coordination Protocol (file-based subagent communication) - **O5**: Integration & Subagent Spawning (OrchestrationManager) ### New Modules - `src/ralph_orchestrator/orchestration/` - 5 modules for subagent coordination - `src/ralph_orchestrator/onboarding/` - 6 modules for project onboarding - `src/ralph_orchestrator/daemon/` - Background daemon mode - `src/ralph_orchestrator/run_manager.py` - Run isolation with unique IDs ### Mobile App (ralph-mobile/) - Expo React Native app for iOS/Android - API layer: 100% complete with all endpoints - Helper layer: 100% complete with validation - UI layer: ~30% complete (Dashboard works, other screens stubbed) - Tests: 13 test files (mocked) ### Test Coverage - **1696 total tests** (up from ~250 baseline) - 105 orchestration-specific tests - ~350 onboarding tests - All passing ### Documentation - Architecture design docs - Validation evidence for all phases - Comprehensive acceptance criteria ## Test plan - [x] All 1696 tests pass (`uv run pytest`) - [x] TUI launches correctly - [x] Daemon mode works (start/stop/status) - [x] REST API endpoints respond - [x] Orchestration module imports correctly - [ ] Mobile app builds (Expo) - *manual verification* --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>