🏗️ Architecture: Should Session 128 focus on Tauri Desktop App MVP (as origin...

Date: 2026-02-20
Time: 04:32:00 UTC
Session: gspace1-128.2
Agent: backend
Confidence: 92%
Status: ✅ Documented

📋 Decision Context

Question:

Should Session 128 focus on Tauri Desktop App MVP (as original plan) or reprioritize to self-extending skills + proactive channels + sanitizeOutput?

Task: Session 128: Road to Beat OpenClaw plan review and enhancement

🎯 Decision Made

Chosen:

Reprioritize: self-extending skills (close OpenClaw's #1 advantage), proactive channel messaging (90% done, massive user value), sanitizeOutput wiring (security gap since S116). Defer Tauri to S131+.

Options Considered:

1. Tauri Desktop App MVP (original plan)
2. Self-extending skills + proactive channels + sanitizeOutput (revised)
3. All four features in one session

💭 Reasoning

Desktop presence already exists (VSCode extension v0.9.5, Chrome extension v0.2.9, desktop daemon POC). Tauri has massive ops overhead (Rust toolchain, code signing, auto-update infra) for marginal user benefit. Self-modification is OpenClaw's killer feature and the skill infrastructure is already excellent — just needs generateSkill() + teach intent + skill discovery wiring. Proactive channels need only ~50 lines of bridge code. sanitizeOutput is a 30-min security fix that's been deferred since S116.

📁 Files Involved

• apis/backend-vitto-api/src/app/engine/skill-manager.service.ts
• apis/backend-vitto-api/src/app/engine/phase-executor.service.ts
• apis/backend-vitto-api/src/app/engine/agent-loop.service.ts
• apis/backend-vitto-api/src/app/proactive/proactive-monitor.service.ts
• apis/backend-vitto-api/src/app/engine/vitto-engine.service.ts
• docs/english/Plans/road-to-beat-openClaw.plan.md

🔗 Related

• Decision ID: dev_1771561918096_9qsjqcc
• Session: 128
• Workspace: gspace1

Auto-generated by MCP Smart Consolidation