clawdie/clawdie-ai
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions
clawdie-ai/MEMORY.md
Clawdie AI b4ce63b630 docs: add ARCHITECTURE.md, refresh MEMORY.md, fix CONTRIBUTING.md identity paths (Sam & Claude) 
- ARCHITECTURE.md: single-entry architecture overview covering host layout,
  agent system, split-brain DB, jail isolation, configuration, infra-as-code,
  channels, and documentation map
- MEMORY.md: update system description to current reality (PG18, split-brain
  3-DB, Phase 7 done, Mevy installed), remove stale open items (WhatsApp,
  Supabase, Docker replacement), update identity file paths
- CONTRIBUTING.md: fix identity file references to match actual paths

Build: pass | Tests: not run (Linux)
2026-04-16 11:58:55 +02:00

3.1 KiB
Raw Blame History

Long-Term Memory

Last updated: 16.Apr.2026

Current System (as of 16.Apr.2026)

  • Host: FreeBSD 15.0 bare-metal, ZFS, Bastille jails, 12 GB RAM
  • Agent runtime: TypeScript (Node.js 24), compiled via tsc to dist/
  • Databases: PostgreSQL 18 + pgvector, split-brain architecture (3 DBs per agent)
    • {agent}_ops — tasks, agents, activity, budgets, approvals
    • {agent}_skills — preloaded knowledge chunks (read-only)
    • {agent}_memory — user/agent dynamic memory with embeddings
    • Runs on host (DB_RUNTIME=host) or in db jail (DB_RUNTIME=jail)
  • LLM: Multiple providers via PI_TUI_PROVIDER (zai, openai, local llama-cpp)
  • Control Plane: Multi-agent orchestrator (port 3100)
    • Roles: Orchestrator (80% budget), Sysadmin (10%), DBA (5%), Git Admin (5%)
    • Identity files in .agent/identities/ (SYSADMIN.md, DB_ADMIN.md, GIT_ADMIN.md, COORDINATOR.md)
    • 46+ operational skills in .agent/skills/
  • Channels: Telegram (grammy bot), HTTP REST API
  • Jail isolation: Phase 7 complete (behind CONTROLPLANE_JAIL_ISOLATION feature flag)
  • Second agent: Mevy installed on same host, validates multi-agent + jail isolation

About Sam

  • Name: Samo Blatnik — call him Sam
  • Location: Near Ljubljana, Slovenia (Europe/Ljubljana, CET/CEST)
  • Languages: Slovenian (native), Croatian/Serbian, English (good)
  • Background: IT support (hardware, printers, general troubleshooting)
  • Telegram ID: 85126311
  • Dog: Bucko
  • CV: https://samob.netlify.app/en/

Key Facts

  • SSH key authentication; ports 80, 443, 22
  • Default subnet: 10.0.0.0/24 (WARDEN_SUBNET_BASE=10.0.0), gateway .1
  • Jail slots: .2 reserved, .3 db, .4 cms, .5 llama-cpp, .6 git, .101+ workers, .211-213 agent workers
  • Install: npm run install (20-step orchestrator), just CLI front door (35+ recipes)
  • Docs: markdown source in docs/public/, auto-compiled to HTML and deployed daily
  • Remotes: git@codeberg.org:Clawdie/Clawdie-AI.git (primary), Clawdie-ISO is cross-repo

Open Work Items

  1. Dashboard — Agent-generated dashboards (Phase C/D) need FreeBSD testing
  2. Skills artifact V1 — Bootstrap artifact.sql when docs stabilize
  3. CONTROLPLANE_JAIL_ISOLATION=YES — Flip flag on Mevy, validate Phase 7e end-to-end
  4. Deploy to ai.clawdie.si — DNS + nginx vhost for public API

Historical Context (Feb–Mar 2026, old system)

The sessions below predate the FreeBSD migration. They ran on a Debian/Hostinger VPS using Docker, OpenClaw v1/v2, and cloud LLMs (GLM, OpenRouter). That infrastructure is retired. These memories are kept for continuity — treat infra references in them as historical, not current.

  • [24.02.] Major session: 3 skills created, cost optimisation, assistant renamed to Clawdie, Telegram integration
  • [14.02.] PRD updates v0.3: SSH, Tailscale VPN, user 'clawdie', GLM-5 model configured
  • [11.02.] Collaborative tmux + Claude Code workflow established
  • [10.02.] Strapi CMS deployment planning
  • [09.02.] Memory persistence explained — context files are the persistent layer
  • [06.02.] Initial setup: Sam configured Clawdie as AI assistant, sudo access, fail2ban, cron