Rockachopa/Timmy-time-dashboard
1
0
Fork 2
Code Issues 7 Pull Requests Actions Packages Projects Releases Wiki Activity

[kimi] OpenClaw architecture and deployment research report (#721) #788

Merged
kimi merged 1 commits from kimi/issue-721 into main 2026-03-21 20:36:24 +00:00
Conversation 0 Commits 1 Files Changed 1 +912
kimi commented 2026-03-21 20:35:40 +00:00
Collaborator
Copy Link

Fixes #721

Summary

This PR adds a comprehensive research report on OpenClaw architecture, deployment modes, and Ollama integration for the Timmy Time Dashboard project.

Research Coverage

1. Architecture Overview

  • Core components: Gateway (port 18789), Pi Agent, SOUL.md, AGENTS.md, workspace, sessions
  • Hub-and-spoke architecture diagram
  • Agent Runner execution flow (8 stages)

2. Deployment Modes

  • npm global (recommended for quick start)
  • Docker/Docker Compose (recommended for production)
  • Bare metal/systemd (maximum performance)
  • Specific recommendations for 2GB RAM VPS

3. Ollama Integration

  • Configuration via native /api/chat endpoint
  • Critical finding: 64K token context window minimum required
  • Compatible models (llama3.1, qwen2.5-coder, deepseek-coder-v2)
  • Models for small VPS (≤8B parameters)

4. OpenRouter Configuration

  • Fallback strategy for reliability
  • Free/cheap tier models (gemma-3-4b-it:free, llama-3.1-8b-instruct:free)
  • Hybrid configuration recommendation

5. Hardware Constraints

  • 2GB VPS: NOT viable for local LLMs; external APIs only
  • 4GB VPS: Minimum for text-only agents
  • 8GB VPS: Practical minimum for Ollama + browser automation
  • 4-bit quantization analysis for Qwen 2.5 7B

6. Security

  • Port 18789: NEVER expose to internet
  • Tailscale integration for zero-trust access
  • SSH tunnel alternative
  • Firewall (UFW) configuration
  • Authentication and token management

7. MCP Tools

  • Model Context Protocol overview
  • MCP vs OpenClaw Skills comparison
  • MCPorter bridge tool usage

Key Recommendations for Timmy

  1. For Hermes VPS (2GB RAM): Use external API only (OpenRouter free tier)
  2. Do NOT run Ollama on 2GB VPS - insufficient RAM
  3. Use Tailscale for secure remote access to gateway
  4. Configure dmPolicy: pairing for Telegram channel security
  5. Enable 4GB swap to prevent OOM kills

Document Location

  • docs/research/openclaw-architecture-deployment-guide.md (912 lines, ~31KB)

Testing

  • tox -e format (passed)
  • tox -e unit (2692 passed, 1 skipped)
  • tox -e lint (passed)
Fixes #721 ## Summary This PR adds a comprehensive research report on OpenClaw architecture, deployment modes, and Ollama integration for the Timmy Time Dashboard project. ## Research Coverage ### 1. Architecture Overview - Core components: Gateway (port 18789), Pi Agent, SOUL.md, AGENTS.md, workspace, sessions - Hub-and-spoke architecture diagram - Agent Runner execution flow (8 stages) ### 2. Deployment Modes - npm global (recommended for quick start) - Docker/Docker Compose (recommended for production) - Bare metal/systemd (maximum performance) - Specific recommendations for 2GB RAM VPS ### 3. Ollama Integration - Configuration via native `/api/chat` endpoint - **Critical finding:** 64K token context window minimum required - Compatible models (llama3.1, qwen2.5-coder, deepseek-coder-v2) - Models for small VPS (≤8B parameters) ### 4. OpenRouter Configuration - Fallback strategy for reliability - Free/cheap tier models (gemma-3-4b-it:free, llama-3.1-8b-instruct:free) - Hybrid configuration recommendation ### 5. Hardware Constraints - **2GB VPS:** NOT viable for local LLMs; external APIs only - **4GB VPS:** Minimum for text-only agents - **8GB VPS:** Practical minimum for Ollama + browser automation - 4-bit quantization analysis for Qwen 2.5 7B ### 6. Security - Port 18789: NEVER expose to internet - Tailscale integration for zero-trust access - SSH tunnel alternative - Firewall (UFW) configuration - Authentication and token management ### 7. MCP Tools - Model Context Protocol overview - MCP vs OpenClaw Skills comparison - MCPorter bridge tool usage ## Key Recommendations for Timmy 1. **For Hermes VPS (2GB RAM):** Use external API only (OpenRouter free tier) 2. **Do NOT run Ollama** on 2GB VPS - insufficient RAM 3. Use **Tailscale** for secure remote access to gateway 4. Configure **dmPolicy: pairing** for Telegram channel security 5. Enable **4GB swap** to prevent OOM kills ## Document Location - `docs/research/openclaw-architecture-deployment-guide.md` (912 lines, ~31KB) ## Testing - [x] tox -e format (passed) - [x] tox -e unit (2692 passed, 1 skipped) - [x] tox -e lint (passed)
kimi added 1 commit 2026-03-21 20:35:41 +00:00
docs: Add comprehensive OpenClaw architecture and deployment research report
Some checks failed
Tests / lint (pull_request) Has been cancelled
Details
Tests / test (pull_request) Has been cancelled
Details
b45c990209 
Fixes #721

This research report covers:
- OpenClaw architecture overview (Gateway, Pi agent, SOUL.md, AGENTS.md)
- Deployment modes comparison (npm, Docker, bare metal)
- Ollama integration with context window requirements (64K minimum)
- OpenRouter configuration for fallback and free-tier models
- Hardware constraints analysis for 2GB RAM VPS
- Security best practices (Tailscale, port 18789, authentication)
- MCP tools ecosystem overview
- Specific recommendations for Timmy Time Dashboard deployment
kimi merged commit 19367d6e41 into main 2026-03-21 20:36:24 +00:00
kimi deleted branch kimi/issue-721 2026-03-21 20:36:24 +00:00
Sign in to join this conversation.
Reviewers
No Reviewers
Labels
Clear labels
222-epic
Workshop: Timmy as Presence (Epic #222)
 actionable
Has a concrete code/config task extracted
 assigned-claude
Issue currently assigned to Claude agent — do not assign to another agent
 assigned-gemini
Issue currently assigned to Gemini agent — do not assign to another agent
 assigned-groq
assigned-kimi
Issue currently assigned to Kimi agent — do not assign to another agent
 assigned-manus
Issue currently assigned to Manus agent — do not assign to another agent
 claude-ready
consolidation
Part of a consolidation epic
 deprioritized
Keep open but not blocking P0 work
 deprioritized
Keep open but not blocking P0 work
 duplicate
Duplicate of another issue
 gemini-review
Auto-generated by Gemini, needs relevance review
 groq-ready
harness
Core product: agent framework, heartbeat, inference, memory
 heartbeat
Harness: Agent heartbeat loop
 inference
Harness: Inference and model routing
 infrastructure
Supporting stage: dashboard, CI/CD, deployment, DNS
 kimi-ready
Scoped and ready for Kimi to pick up
 memory-session
Harness: Memory and session crystallization
 morrowind
Harness: Morrowind embodiment
 needs-design
Needs architectural design before implementation
 needs-extraction
Philosophy with unextracted engineering work
 p0-critical
Priority 0: Must fix now
 p1-important
Priority 1: Important, next sprint
 p2-backlog
Priority 2: Backlog, do when time permits
 philosophy
Philosophical foundation — informs architecture decisions
 rejected-direction
Closed: rejected or superseded direction
 seed:know-purpose
Three Seeds: KNOW YOUR PURPOSE
 seed:serve-real
Three Seeds: SERVE THE REAL
 seed:tell-truth
Three Seeds: TELL THE TRUTH
 sovereignty
Harness: Sovereignty stack
No Label
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
Timmy claude grok groq kimi manus perplexity Rockachopa
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: Rockachopa/Timmy-time-dashboard#788
Delete Branch "kimi/issue-721"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?