feat: v0.1.65

CHANGELOG.md

@@ -9,6 +9,9 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## Recent Releases
 
+**v0.1.65 (March 18, 2026)** - MassGen Refinery Plugin
+Standalone MCP servers (quality, workflow, media) bring MassGen's checklist-based evaluation and multi-round refinement to Claude Code through the massgen-refinery plugin. Single-agent refinement working; multi-agent experimental.
+
 **v0.1.64 (March 16, 2026)** - Gemini CLI Backend
 Gemini CLI as a first-class backend with session persistence, MCP tools, and Docker support. WebSocket streaming for OpenAI Response API. Execution trace analyzer subagent. Copilot Docker mode. Response API duplicate fix.
 
@@ -29,6 +32,18 @@ Planning improvements with auto-added improvements to task plan and plan review
 
 ---
 
+## [0.1.65] - 2026-03-18
+
+### Added
+- **Quality Server** ([#1007](https://github.com/massgen/MassGen/pull/1007)): Standalone `massgen_quality_tools` MCP server with session-based checklist evaluation, configurable scoring thresholds, improvement proposals, and coverage validation
+- **Workflow Server** ([#1007](https://github.com/massgen/MassGen/pull/1007)): Standalone `massgen_workflow_tools` MCP server with multi-round answer submission, automatic deliverable snapshots, and vote support
+- **Media Server** ([#1007](https://github.com/massgen/MassGen/pull/1007)): Standalone `massgen_media_tools` MCP server with image/video/audio generation and critical-first media analysis
+
+### Technical Details
+- **Major Focus**: MassGen Refinery Plugin — standalone MCP servers for Claude Code
+- **PRs Merged**: [#1007](https://github.com/massgen/MassGen/pull/1007)
+- **Contributors**: @ncrispino, @HenryQi and the MassGen team
+
 ## [0.1.64] - 2026-03-16
 
 ### Added
@@ -43,7 +58,7 @@ Planning improvements with auto-added improvements to task plan and plan review
 ### Technical Details
 - **Major Focus**: Gemini CLI Backend
 - **PRs Merged**: [#999](https://github.com/massgen/MassGen/pull/999), [#990](https://github.com/massgen/MassGen/pull/990), [#1002](https://github.com/massgen/MassGen/pull/1002), [#1000](https://github.com/massgen/MassGen/pull/1000)
-- **Contributors**: @ncrispino, @HenryQi, @db-ol and the MassGen team
+- **Contributors**: @praneeth999, @ncrispino, @HenryQi, @db-ol and the MassGen team
 
 ## [0.1.63] - 2026-03-13
 

CONTRIBUTING.md

@@ -359,7 +359,7 @@ Create a `.env` file in the `massgen` directory as described in [README](README.
 
 ## 🔧 Development Workflow
 
-> **Important**: Our next version is v0.1.65. If you want to contribute, please contribute to the `dev/v0.1.65` branch (or `main` if dev/v0.1.65 doesn't exist yet).
+> **Important**: Our next version is v0.1.66. If you want to contribute, please contribute to the `dev/v0.1.66` branch (or `main` if dev/v0.1.66 doesn't exist yet).
 
 ### 1. Create Feature Branch
 
@@ -368,7 +368,7 @@ Create a `.env` file in the `massgen` directory as described in [README](README.
 git fetch upstream
 
 # Create feature branch from dev/v0.1.60 (or main if dev branch doesn't exist yet)
-git checkout -b feature/your-feature-name upstream/dev/v0.1.65
+git checkout -b feature/your-feature-name upstream/dev/v0.1.66
 ```
 
 ### 2. Make Your Changes
@@ -507,7 +507,7 @@ git push origin feature/your-feature-name
 ```
 
 Then create a pull request on GitHub:
-- Base branch: `dev/v0.1.65` (or `main` if dev branch doesn't exist yet)
+- Base branch: `dev/v0.1.66` (or `main` if dev branch doesn't exist yet)
 - Compare branch: `feature/your-feature-name`
 - Add clear description of changes
 - Link any related issues
@@ -617,7 +617,7 @@ Have a significant feature idea not covered by existing tracks?
 - [ ] Tests pass locally
 - [ ] Documentation is updated if needed
 - [ ] Commit messages follow convention
-- [ ] PR targets `dev/v0.1.65` branch (or `main` if dev branch doesn't exist yet)
+- [ ] PR targets `dev/v0.1.66` branch (or `main` if dev branch doesn't exist yet)
 
 ### PR Description Should Include
 

README.md

@@ -69,7 +69,7 @@ This project started with the "threads of thought" and "iterative refinement" id
 <details open>
 <summary><h3>🆕 Latest Features</h3></summary>
 
-- [v0.1.64 Features](#-latest-features-v0164)
+- [v0.1.65 Features](#-latest-features-v0165)
 </details>
 
 <details open>
@@ -122,15 +122,15 @@ This project started with the "threads of thought" and "iterative refinement" id
 <details open>
 <summary><h3>🗺️ Roadmap</h3></summary>
 
-- [Recent Achievements (v0.1.64)](#recent-achievements-v0164)
-- [Previous Achievements (v0.0.3 - v0.1.63)](#previous-achievements-v003---v0163)
+- [Recent Achievements (v0.1.65)](#recent-achievements-v0165)
+- [Previous Achievements (v0.0.3 - v0.1.64)](#previous-achievements-v003---v0164)
 - [Key Future Enhancements](#key-future-enhancements)
   - Bug Fixes & Backend Improvements
   - Advanced Agent Collaboration
   - Expanded Model, Tool & Agent Integrations
   - Improved Performance & Scalability
   - Enhanced Developer Experience
-- [v0.1.65 Roadmap](#v0165-roadmap)
+- [v0.1.66 Roadmap](#v0166-roadmap)
 </details>
 
 <details open>
@@ -155,20 +155,20 @@ This project started with the "threads of thought" and "iterative refinement" id
 
 ---
 
-## 🆕 Latest Features (v0.1.64)
+## 🆕 Latest Features (v0.1.65)
 
-**🎉 Released: March 16, 2026**
+**🎉 Released: March 18, 2026**
 
-**What's New in v0.1.64:**
-- **🔌 Gemini CLI Backend** - Google's Gemini CLI as a first-class backend with session persistence, MCP tools, and Docker support.
-- **⚡ WebSocket Streaming** - Persistent WebSocket transport for OpenAI Response API with auto-reconnection.
-- **🔍 Execution Trace Analyzer** - New subagent type for mechanistic analysis of agent execution traces.
+**What's New in v0.1.65:**
+- **🔧 MassGen Refinery Plugin** - Standalone MCP servers (quality, workflow, media) bring MassGen's checklist-based evaluation to Claude Code. Single-agent refinement working; multi-agent experimental.
+- **✅ Quality Server** - Session-based checklist evaluation with scoring thresholds and improvement proposals.
+- **📋 Workflow Server** - Multi-round answer submission with automatic deliverable snapshots.
 
-**Try v0.1.64 Features:**
+**Try v0.1.65 Features:**
 ```bash
-pip install massgen==0.1.64
-# Try the Gemini CLI backend
-uv run massgen --config @examples/providers/gemini/gemini_cli_local "Explain quantum computing"
+pip install massgen==0.1.65
+# The standalone MCP servers are available for the massgen-refinery Claude Code plugin
+# https://github.com/massgen/massgen-refinery
 ```
 
 → [See full release history and examples](massgen/configs/README.md#release-history--examples)
@@ -1240,23 +1240,18 @@ MassGen is currently in its foundational stage, with a focus on parallel, asynch
 
 ⚠️ **Early Stage Notice:** As MassGen is in active development, please expect upcoming breaking architecture changes as we continue to refine and improve the system.
 
-### Recent Achievements (v0.1.64)
+### Recent Achievements (v0.1.65)
 
-**🎉 Released: March 16, 2026**
+**🎉 Released: March 18, 2026**
 
-#### Gemini CLI Backend
-- **Gemini CLI** ([#999](https://github.com/massgen/MassGen/pull/999), [#952](https://github.com/massgen/MassGen/issues/952)): New subprocess-based backend for Google's Gemini CLI with session persistence, MCP tools, and Docker support
-- **Native Hook Adapter**: Standardized tool execution for Gemini CLI via hook-based IPC
+#### MassGen Refinery Plugin
+- **Quality Server** ([#1007](https://github.com/massgen/MassGen/pull/1007)): Standalone `massgen_quality_tools` MCP server with session-based checklist evaluation, scoring thresholds, and improvement proposals
+- **Workflow Server** ([#1007](https://github.com/massgen/MassGen/pull/1007)): Standalone `massgen_workflow_tools` MCP server with multi-round answer submission and deliverable snapshots
+- **Media Server** ([#1007](https://github.com/massgen/MassGen/pull/1007)): Standalone `massgen_media_tools` MCP server with image/video/audio generation and media analysis
 
-#### WebSocket & Trace Analyzer
-- **WebSocket Mode** ([#990](https://github.com/massgen/MassGen/pull/990)): Persistent WebSocket transport for OpenAI Response API with auto-reconnection
-- **Execution Trace Analyzer** ([#1002](https://github.com/massgen/MassGen/pull/1002)): New subagent type for mechanistic analysis of agent execution traces with 7-dimension evaluation
+### Previous Achievements (v0.0.3 - v0.1.64)
 
-#### Copilot Docker & Fixes
-- **Copilot Docker Mode** ([#999](https://github.com/massgen/MassGen/pull/999)): Containerized tool execution for Copilot backend
-- **Response API Fix** ([#1000](https://github.com/massgen/MassGen/pull/1000)): Prevent duplicate item errors in recursive tool loops
-
-### Previous Achievements (v0.0.3 - v0.1.63)
+✅ **Gemini CLI Backend (v0.1.64)**: Gemini CLI as a first-class backend with session persistence, MCP tools, and Docker support. WebSocket streaming for OpenAI Response API. Execution trace analyzer subagent. Copilot Docker mode.
 
 ✅ **Ensemble & Contracts (v0.1.63)**: Subagent ensemble pattern with `disable_injection` and `defer_voting_until_all_answered` as defaults. Round evaluator transformation pressure and success contracts. Lighter refinement for subagents. Killed agent handling.
 
@@ -1529,9 +1524,9 @@ MassGen is currently in its foundational stage, with a focus on parallel, asynch
 
 We welcome community contributions to achieve these goals.
 
-### v0.1.65 Roadmap
+### v0.1.66 Roadmap
 
-Version 0.1.65 focuses on cloud execution:
+Version 0.1.66 focuses on cloud execution:
 
 #### Planned Features
 - **Cloud Modal MVP** ([#982](https://github.com/massgen/MassGen/issues/982)): Run MassGen as a cloud job on Modal

README_PYPI.md

@@ -68,7 +68,7 @@ This project started with the "threads of thought" and "iterative refinement" id
 <details open>
 <summary><h3>🆕 Latest Features</h3></summary>
 
-- [v0.1.64 Features](#-latest-features-v0164)
+- [v0.1.65 Features](#-latest-features-v0165)
 </details>
 
 <details open>
@@ -121,15 +121,15 @@ This project started with the "threads of thought" and "iterative refinement" id
 <details open>
 <summary><h3>🗺️ Roadmap</h3></summary>
 
-- [Recent Achievements (v0.1.64)](#recent-achievements-v0164)
-- [Previous Achievements (v0.0.3 - v0.1.63)](#previous-achievements-v003---v0163)
+- [Recent Achievements (v0.1.65)](#recent-achievements-v0165)
+- [Previous Achievements (v0.0.3 - v0.1.64)](#previous-achievements-v003---v0164)
 - [Key Future Enhancements](#key-future-enhancements)
   - Bug Fixes & Backend Improvements
   - Advanced Agent Collaboration
   - Expanded Model, Tool & Agent Integrations
   - Improved Performance & Scalability
   - Enhanced Developer Experience
-- [v0.1.65 Roadmap](#v0165-roadmap)
+- [v0.1.66 Roadmap](#v0166-roadmap)
 </details>
 
 <details open>
@@ -154,20 +154,20 @@ This project started with the "threads of thought" and "iterative refinement" id
 
 ---
 
-## 🆕 Latest Features (v0.1.64)
+## 🆕 Latest Features (v0.1.65)
 
-**🎉 Released: March 16, 2026**
+**🎉 Released: March 18, 2026**
 
-**What's New in v0.1.64:**
-- **🔌 Gemini CLI Backend** - Google's Gemini CLI as a first-class backend with session persistence, MCP tools, and Docker support.
-- **⚡ WebSocket Streaming** - Persistent WebSocket transport for OpenAI Response API with auto-reconnection.
-- **🔍 Execution Trace Analyzer** - New subagent type for mechanistic analysis of agent execution traces.
+**What's New in v0.1.65:**
+- **🔧 MassGen Refinery Plugin** - Standalone MCP servers (quality, workflow, media) bring MassGen's checklist-based evaluation to Claude Code. Single-agent refinement working; multi-agent experimental.
+- **✅ Quality Server** - Session-based checklist evaluation with scoring thresholds and improvement proposals.
+- **📋 Workflow Server** - Multi-round answer submission with automatic deliverable snapshots.
 
-**Try v0.1.64 Features:**
+**Try v0.1.65 Features:**
 ```bash
-pip install massgen==0.1.64
-# Try the Gemini CLI backend
-uv run massgen --config @examples/providers/gemini/gemini_cli_local "Explain quantum computing"
+pip install massgen==0.1.65
+# The standalone MCP servers are available for the massgen-refinery Claude Code plugin
+# https://github.com/massgen/massgen-refinery
 ```
 
 → [See full release history and examples](massgen/configs/README.md#release-history--examples)
@@ -1239,23 +1239,18 @@ MassGen is currently in its foundational stage, with a focus on parallel, asynch
 
 ⚠️ **Early Stage Notice:** As MassGen is in active development, please expect upcoming breaking architecture changes as we continue to refine and improve the system.
 
-### Recent Achievements (v0.1.64)
+### Recent Achievements (v0.1.65)
 
-**🎉 Released: March 16, 2026**
+**🎉 Released: March 18, 2026**
 
-#### Gemini CLI Backend
-- **Gemini CLI** ([#999](https://github.com/massgen/MassGen/pull/999), [#952](https://github.com/massgen/MassGen/issues/952)): New subprocess-based backend for Google's Gemini CLI with session persistence, MCP tools, and Docker support
-- **Native Hook Adapter**: Standardized tool execution for Gemini CLI via hook-based IPC
+#### MassGen Refinery Plugin
+- **Quality Server** ([#1007](https://github.com/massgen/MassGen/pull/1007)): Standalone `massgen_quality_tools` MCP server with session-based checklist evaluation, scoring thresholds, and improvement proposals
+- **Workflow Server** ([#1007](https://github.com/massgen/MassGen/pull/1007)): Standalone `massgen_workflow_tools` MCP server with multi-round answer submission and deliverable snapshots
+- **Media Server** ([#1007](https://github.com/massgen/MassGen/pull/1007)): Standalone `massgen_media_tools` MCP server with image/video/audio generation and media analysis
 
-#### WebSocket & Trace Analyzer
-- **WebSocket Mode** ([#990](https://github.com/massgen/MassGen/pull/990)): Persistent WebSocket transport for OpenAI Response API with auto-reconnection
-- **Execution Trace Analyzer** ([#1002](https://github.com/massgen/MassGen/pull/1002)): New subagent type for mechanistic analysis of agent execution traces with 7-dimension evaluation
+### Previous Achievements (v0.0.3 - v0.1.64)
 
-#### Copilot Docker & Fixes
-- **Copilot Docker Mode** ([#999](https://github.com/massgen/MassGen/pull/999)): Containerized tool execution for Copilot backend
-- **Response API Fix** ([#1000](https://github.com/massgen/MassGen/pull/1000)): Prevent duplicate item errors in recursive tool loops
-
-### Previous Achievements (v0.0.3 - v0.1.63)
+✅ **Gemini CLI Backend (v0.1.64)**: Gemini CLI as a first-class backend with session persistence, MCP tools, and Docker support. WebSocket streaming for OpenAI Response API. Execution trace analyzer subagent. Copilot Docker mode.
 
 ✅ **Ensemble & Contracts (v0.1.63)**: Subagent ensemble pattern with `disable_injection` and `defer_voting_until_all_answered` as defaults. Round evaluator transformation pressure and success contracts. Lighter refinement for subagents. Killed agent handling.
 
@@ -1528,9 +1523,9 @@ MassGen is currently in its foundational stage, with a focus on parallel, asynch
 
 We welcome community contributions to achieve these goals.
 
-### v0.1.65 Roadmap
+### v0.1.66 Roadmap
 
-Version 0.1.65 focuses on cloud execution:
+Version 0.1.66 focuses on cloud execution:
 
 #### Planned Features
 - **Cloud Modal MVP** ([#982](https://github.com/massgen/MassGen/issues/982)): Run MassGen as a cloud job on Modal

ROADMAP.md

@@ -1,10 +1,10 @@
 # MassGen Roadmap
 
-**Current Version:** v0.1.64
+**Current Version:** v0.1.65
 
 **Release Schedule:** Mondays, Wednesdays, Fridays @ 9am PT
 
-**Last Updated:** March 16, 2026
+**Last Updated:** March 18, 2026
 
 This roadmap outlines MassGen's development priorities for upcoming releases. Each release focuses on specific capabilities with real-world use cases.
 
@@ -42,14 +42,25 @@ Want to contribute or collaborate on a specific track? Reach out to the track ow
 
 | Release | Target | Feature | Owner | Use Case |
 |---------|--------|---------|-------|----------|
-| **v0.1.65** | 03/18/26 | Cloud Modal MVP | @ncrispino | Run MassGen as a cloud job on Modal ([#982](https://github.com/massgen/MassGen/issues/982)) |
-| **v0.1.66** | 03/20/26 | OpenAI Audio API | @ncrispino | Support OpenAI audio API for audio understanding ([#960](https://github.com/massgen/MassGen/issues/960)) |
-| **v0.1.67** | 03/23/26 | Image/Video Edit Capabilities | @ncrispino | Check and support img/video editing capabilities ([#959](https://github.com/massgen/MassGen/issues/959)) |
+| **v0.1.66** | 03/20/26 | Cloud Modal MVP | @ncrispino | Run MassGen as a cloud job on Modal ([#982](https://github.com/massgen/MassGen/issues/982)) |
+| **v0.1.67** | 03/23/26 | OpenAI Audio API | @ncrispino | Support OpenAI audio API for audio understanding ([#960](https://github.com/massgen/MassGen/issues/960)) |
+| **v0.1.68** | 03/25/26 | Image/Video Edit Capabilities | @ncrispino | Check and support img/video editing capabilities ([#959](https://github.com/massgen/MassGen/issues/959)) |
 
 *All releases ship on MWF @ 9am PT when ready*
 
 ---
 
+## ✅ v0.1.65 - MassGen Refinery Plugin (Completed)
+
+**Released:** March 18, 2026 | PR: [#1007](https://github.com/massgen/MassGen/pull/1007)
+
+### Features
+- **Quality Server**: Standalone `massgen_quality_tools` MCP server with session-based checklist evaluation, scoring thresholds, and improvement proposals
+- **Workflow Server**: Standalone `massgen_workflow_tools` MCP server with multi-round answer submission and deliverable snapshots
+- **Media Server**: Standalone `massgen_media_tools` MCP server with image/video/audio generation and media analysis
+
+---
+
 ## ✅ v0.1.64 - Gemini CLI Backend (Completed)
 
 **Released:** March 16, 2026 | PRs: [#999](https://github.com/massgen/MassGen/pull/999), [#990](https://github.com/massgen/MassGen/pull/990), [#1002](https://github.com/massgen/MassGen/pull/1002), [#1000](https://github.com/massgen/MassGen/pull/1000)
@@ -63,7 +74,7 @@ Want to contribute or collaborate on a specific track? Reach out to the track ow
 
 ---
 
-## 📋 v0.1.65 - Cloud Modal MVP
+## 📋 v0.1.66 - Cloud Modal MVP
 
 ### Features
 
@@ -79,7 +90,7 @@ Want to contribute or collaborate on a specific track? Reach out to the track ow
 
 ---
 
-## 📋 v0.1.66 - OpenAI Audio API
+## 📋 v0.1.67 - OpenAI Audio API
 
 ### Features
 
@@ -95,7 +106,7 @@ Want to contribute or collaborate on a specific track? Reach out to the track ow
 
 ---
 
-## 📋 v0.1.67 - Image/Video Edit Capabilities
+## 📋 v0.1.68 - Image/Video Edit Capabilities
 
 ### Features
 

ROADMAP_v0.1.65.md → ROADMAP_v0.1.66.md

@@ -1,10 +1,10 @@
-# MassGen v0.1.65 Roadmap
+# MassGen v0.1.66 Roadmap
 
-**Target Release:** March 18, 2026
+**Target Release:** March 20, 2026
 
 ## Overview
 
-Version 0.1.65 focuses on running MassGen as a cloud job on Modal.
+Version 0.1.66 focuses on running MassGen as a cloud job on Modal.
 
 ---
 
@@ -27,5 +27,5 @@ Version 0.1.65 focuses on running MassGen as a cloud job on Modal.
 
 ## Related Tracks
 
-- **v0.1.64**: Gemini CLI Backend — Gemini CLI, WebSocket streaming, execution trace analyzer, Copilot Docker ([#999](https://github.com/massgen/MassGen/pull/999), [#990](https://github.com/massgen/MassGen/pull/990), [#1002](https://github.com/massgen/MassGen/pull/1002))
-- **v0.1.66**: OpenAI Audio API ([#960](https://github.com/massgen/MassGen/issues/960))
+- **v0.1.65**: MassGen Refinery Plugin — standalone MCP servers for Claude Code ([#1007](https://github.com/massgen/MassGen/pull/1007))
+- **v0.1.67**: OpenAI Audio API ([#960](https://github.com/massgen/MassGen/issues/960))