Manage Hermes Agent gateway lifecycle, platform connections, and WSL troubleshooting.
npx skills add https://github.com/tapway/shogun-os --skill hermes-agent-tapway
Please help me install this Agent Skill. Skill: hermes-agent Source: https://github.com/tapway/shogun-os/tree/main/skills/hermes/hermes-agent Command: npx skills add https://github.com/tapway/shogun-os --skill hermes-agent-tapway
hermes gateway restart commands. - Failure Diagnosis: Step-by-step fixes for 14+ documented failure modes including zombie processes, slow LLM responses, raw tool-call XML leaking into Slack, PGLite WAL corruption, and config YAML parse errors. - Auto-Start on WSL: Configure .bashrc and Windows Task Scheduler so the gateway survives VM reboots and terminal closures. - Use Case: Your Slack bot stops responding after a gateway restart. Use this Skill to identify the tmux session exit, install the watchdog wrapper, and restore the gateway with auto-recovery in minutes. ## Quick Start Ask the agent to check the Hermes gateway status and restart it inside a tmux session using the watchdog script.High-intent search queries and answers about installing and using this skill.
Run the gateway inside a detached tmux session with the watchdog script: `tmux new-session -d -s hermes-gateway '~/.local/bin/hermes-gateway-watchdog'`. The tmux server is parented by init, so the gateway survives chat session termination.
The gateway was started as a child of the chat session, so it receives SIGTERM when the session's process group is reaped. Check gateway.log for 'Shutdown context: parent_name=hermes' and restart it inside a dedicated tmux session instead.
A zombie gateway process still holds the token. Find defunct processes with `ps aux | grep hermes`, kill them with `kill -9 <PID>`, clean up stale tmux sessions, then start a fresh gateway in a new tmux session.
Yes, using two layers: a `.bashrc` snippet that starts the tmux watchdog when any terminal opens, plus a Windows Task Scheduler ONLOGON task that calls `wsl.exe` to launch the watchdog without requiring a terminal.
Common causes are `reasoning_effort` forcing chain-of-thought on every message, bloated memory above 90% utilization, and large skill description counts in the system prompt. Set reasoning_effort to empty and add a FAST PATH channel prompt for trivial messages.
Do not use it for cron job management (use hermes-cron-jobs), model or provider configuration (use hermes-model-config), or profile persona authoring (use profile-management). It covers only gateway operations and platform connectivity.
ToqsickConfigure messaging platforms via Hermes gateway with bot creation and token management.
timchapDiagnose and fix Hermes Agent cron job failures including missed runs, delivery errors, and provider drift.
GetFresh-VenturesConfigure and monitor cron-like jobs across macOS launchd, Linux systemd, and cron.
CHENHUI-XOperate deployed Hermes Agent instances covering credentials, messaging gateways, webhooks, and systemd services.