4.0 KiB
4.0 KiB
title, type, status, created, updated, verified_on, confidence, tags, sources
| title | type | status | created | updated | verified_on | confidence | tags | sources | ||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| QwenPaw | system | active | 2026-08-15 | 2026-11-04 | 2026-11-04 | high |
|
|
QwenPaw
Purpose
Multi-agent gateway platform. Runs on CT465 with its own workspace, agent profiles, and model routing. Used for group chats and specialised agent tasks.
Location
- Container: LXC CT465 (
qwenpaw) - Hostname:
qwenpaw - Software: QwenPaw with Paseo daemon
Current State
- Status: Active —
paseo.serviceactive on CT465, re-verified 2026-11-04 - Default chat model: OmniRoute Smart Route (
auto/smart) - Config format: JSON (
agent.jsonper workspace) - Daemon:
paseo.service— system unit on CT465 (distinct from the user unitpaseo-daemon.servicethat runs on CT460; see systems/paseo) - CT465 listening ports (live 2026-11-04):
0.0.0.0:6767(Paseo),*:13500,127.0.0.1:25463,127.0.0.1:11434(Ollama),127.0.0.1:38585,100.110.72.126:42510,:22- ⚠ Corrected 2026-11-04. The old claim that "PVE exporter (
100.96.150.69:9221), a Python service on:9876, and local SMTP:25" were listening on CT465 was wrong on all three counts. Livess -ltninside CT465 shows none of them.:9221and:9876listen on thepvehypervisor host itself (100.96.100.82), not in the guest, and100.96.150.69is CT200 (the Docker host), not the PVE exporter.:25is the hypervisor's own loopback SMTP, also not in CT465.
- ⚠ Corrected 2026-11-04. The old claim that "PVE exporter (
- Port 6767 has two live implementations — do not conflate them: CT465's own
paseo.servicelistens on0.0.0.0:6767and answers HTTP 404 on/(the daemon is alive;/is not a route). The tailnet nodepaseo(100.72.188.93) is a different guest — CT1261 onm93p— also runningpaseo.serviceon:6767, and it answers HTTP 200 on/. CT460's own user-unit daemon is a third instance (also:6767, HTTP 404 on/;/api/health→ 200). - CT400/CT465 identity: confirmed CT465 =
qwenpaw, running
Operating Layers
QwenPaw has several layers that must not be conflated:
- PVE/LXC layer: container status, guest access, resources
- Paseo layer: long-running daemon/supervisor (
paseo.service) - Workspace layer: agent profiles with
agent.jsonandactive_model - Provider layer: remote providers (Kilo/OpenRouter) or local (Ollama)
- Channel layer: Telegram or other messaging adapters
Key Rules
- Do not assume "process is running" means "agent works." Model resolution and a real response are separate checks.
- Verify model resolution per agent. A default agent can work while a specialist agent silently has no model.
- Preserve existing provider/model routes. Do not repin unrelated agents or replace known-good routes.
- Default chat model: Use OmniRoute Smart Route (
auto/smart) per Anthony's preference.
Diagnostics
qwenpaw doctor --deep --timeout 10 --llm-timeout 20
qwenpaw models list
qwenpaw daemon status
Provider Verification
When adding an OpenAI-compatible provider from Proxmox:
- Verify target by VMID and hostname before writing config
- CT600/
omnirouteis the live OmniRoute service at100.88.81.19:20128— verified serving 2026-10-11, still live 2026-11-04 - ⚠ Corrected 2026-11-04: CT601/
omniis no longer present on either hypervisor —pct liston bothpveandm93pshows no CT601 at all, and theomnitailnet peer (100.93.204.4) has dropped out of the tailnet entirely. The guest was removed, not merely stopped. Use CT600. - A third instance
omniroute-1at100.78.63.20:20128is also live; Hermes registers it as provideromni-101 - Query
/v1/modelsusing the exact API key stored for QwenPaw, not a management key - See systems/omniroute for the full live node table
Related
- agents/hermes-production — Production Hermes
- systems/omniroute — Model routing
- systems/model-providers — Provider catalog
- infrastructure/proxmox — PVE host