Skip to main content
Glama

DollhouseMCP

by DollhouseMCP
V1_6_0_RELEASE_PRIORITIES.md5.21 kB
# v1.6.0 Release Priorities **Last Updated**: August 22, 2025 **Current Status**: Planning & Prioritization **QA Success Rate**: 94% → Target 100% ## Executive Summary For v1.6.0 release, we need to focus on: 1. **Critical bug fixes** that block core functionality 2. **QA improvements** to reach 100% success rate 3. **Tool consolidation** to reduce overhead 4. **Security issues** (OAuth exposure, validation bypass) ## QA Analytics Summary ### Current Performance (94% Success Rate) - **15/16 tests passing** in latest run - **42 tools available** and discovered - **Sub-50ms response** for most operations - **Stable memory** usage (73-82MB RSS) ### Top 5 Areas to Fix for 100% Success 1. **`browse_collection` Timeout Issues** (33% failure rate) - Network timeouts at 300ms - Needs configurable timeout & retry logic 2. **Element Activation Validation** - Fails on non-existent elements - Needs pre-validation and better error handling 3. **Tool Discovery Performance** - Highly variable: 6-207ms - Needs caching mechanism 4. **Network Resilience** - No retry mechanisms - Needs exponential backoff 5. **Error Recovery** - Limited graceful degradation - Needs fallback mechanisms ## Critical Issues for v1.6.0 ### 🔴 MUST FIX (Blocking Issues) #### 1. **#404 - CRITICAL: Expose element system through MCP tools** - Element system exists but not accessible via MCP - This is the core feature of v1.6.0 - **Status**: Needs immediate implementation #### 2. **#519 - SECURITY: Rotate OAuth Client ID** - OAuth client ID exposed in git history - Security vulnerability - **Status**: Needs rotation and secure storage #### 3. **#544 - Security validation bypass** - Portfolio content with existing frontmatter bypasses validation - Security vulnerability - **Status**: Needs validation fix #### 4. **#517 - Fix OAuth device flow authentication** - Token not stored after GitHub authorization - Blocks collection submission workflow - **Status**: Needs token persistence fix #### 5. **#610 - Fix race condition in server initialization** - MCP commands fail due to initialization race - Causes intermittent failures - **Status**: Needs initialization sequencing fix ### 🟡 HIGH PRIORITY (Should Have) #### QA & Testing - **#681** - Transition to blocking QA tests in CI - **#670** - QA Framework priority tasks for v1.6.0 - **#598** - Fix E2E roundtrip workflow tests - **#695** - Configurable server startup timeout (from PR #689) #### Tool Consolidation - **#546** - Consolidate MCP tools to reduce overhead - **#632** - Remove UpdateTools (5 auto-update tools) - **#633** - Remove 9 redundant PersonaTools #### Workflow Issues - **#528** - Portfolio uploads all elements to personas directory - **#529** - Collection submission workflow stops after upload - **#530** - OAuth workflow needs better UX explanations ### 🟢 NICE TO HAVE (Can Wait) #### Performance Enhancements - **#698** - Concurrent test execution - **#699** - Performance baseline establishment - **#680** - Performance benchmarking metrics #### Future Features - **#412** - Re-implement Ensemble element system - **#413** - Re-implement Memory element system - **#427** - Comprehensive audit logging system #### Low Priority Improvements - **#696** - Metrics retention policy - **#697** - Network failure simulation tests - **#688** - Optimize QA cleanup with batch operations ## Recommended Action Plan ### Week 1: Critical Fixes 1. **Fix element system MCP exposure (#404)** - Estimated: 2-3 days - This unblocks core v1.6.0 functionality 2. **Fix security issues (#519, #544)** - Estimated: 1 day - Critical for security 3. **Fix OAuth flow (#517)** - Estimated: 1 day - Unblocks submission workflow ### Week 2: QA Improvements 1. **Fix `browse_collection` timeouts** - Add configurable timeout (#695) - Add retry logic with exponential backoff - Estimated: 1 day 2. **Fix race condition (#610)** - Proper initialization sequencing - Estimated: 1 day 3. **Consolidate tools (#546, #632, #633)** - Remove redundant tools - Reduce overhead - Estimated: 2 days ### Week 3: Testing & Polish 1. **E2E test fixes (#598)** 2. **Transition to blocking QA tests (#681)** 3. **Performance optimization based on metrics** ## Success Metrics for v1.6.0 - ✅ **100% QA test success rate** - ✅ **Element system fully accessible via MCP** - ✅ **OAuth flow working end-to-end** - ✅ **No critical security vulnerabilities** - ✅ **Tool count reduced by ~30%** - ✅ **E2E tests passing on all platforms** ## Duplicate Issues to Close Several issues were duplicated (created twice): - Close #690-694 (duplicates of #695-699) - Close #678-682 (earlier QA issues superseded by PR #689) ## Next Steps 1. **Start with #404** - Element system MCP exposure (most critical) 2. **Fix security issues** - #519, #544 (high risk) 3. **Fix OAuth flow** - #517 (blocks workflow) 4. **Improve QA to 100%** - Targeted fixes based on analytics 5. **Tool consolidation** - Reduce from 42 to ~30 tools --- *This prioritization focuses on delivering a stable, secure v1.6.0 with the element system fully functional and accessible.*

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/DollhouseMCP/DollhouseMCP'

If you have feedback or need assistance with the MCP directory API, please join our Discord server