How do I install the Operating Remote skill?

Run npx skillmds@latest add dinglebear-ai/operating-remote in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Operating Remote safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Operating Remote?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Operating Remote free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Operating Remote?

dinglebear-ai (@dinglebear-ai) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Operating Remote by dinglebear-ai — Agent Skill

Name: Operating Remote
Author: dinglebear-ai

Operating Remote

This skill should be used when the user asks to inspect, troubleshoot, change, SSH into, or operate a named remote host that has a Plexus `REMOTE.md` profile, such as example-host, media-server, devbox, or another managed device. Before changing state, load durable host memory and live context with `remote-context.py <host>`. This is host-scoped operational memory for remote machines.

dinglebear-ai 7e77673 4 files · 3.4 KB Updated Sep 17, 2026

File contents

Operating Remote Hosts With Plexus

Use this skill when the task is about a named remote machine and Plexus has a matching REMOTE.md profile. Profiles live in the Plexus data directory: $PLEXUS_DATA_DIR, $CLAUDE_PLUGIN_DATA, $CODEX_PLUGIN_DATA, or ~/.plexus as a fallback.

Required First Step

Identify the host from the user's request and run the context loader before making changes:

python3 "${CLAUDE_PLUGIN_ROOT:-${CODEX_PLUGIN_ROOT:-plugins/plexus}}/scripts/remote-context.py" <host>

Use --no-probe only when the user asks for an offline plan or when SSH/live tools are unavailable. Treat <host> as a profile name, not a shell fragment; use only the exact host token from the request or from known Plexus profiles.

How To Use The Context

Treat REMOTE.md as durable host memory: roles, important paths, access patterns, guardrails, and known quirks. Treat live probe output as the current state. If the two disagree, call out the discrepancy and prefer observed live state for operational decisions.

Operating Rules

Use non-interactive SSH: ssh -o BatchMode=yes <host> <command>.
Inspect before changing state: uptime, disk, memory, failed units, Docker containers, and recent syslog entries.
Prefer host-specific guardrails in REMOTE.md over generic assumptions.
For service work, capture enough before/after evidence to prove the change.
If a task touches a reverse proxy, certificate, firewall, storage pool, or auth boundary, validate first and use a rollback path.

Related Context

Plexus pairs well with cortex:cortex-logs: recent logs and AI/session history explain what changed before the current request. It also composes with service-specific skills like tailscale, unraid, create-swag-config, and homelab-map.

---name: operating-remotedescription: This skill should be used when the user asks to inspect, troubleshoot, change, SSH into, or operate a named remote host that has a Plexus `REMOTE.md` profile, such as example-host, media-server, devbox, or another managed device. Before changing state, load durable host memory and live context with `remote-context.py <host>`. This is host-scoped operational memory for remote machines.---# Operating Remote Hosts With PlexusUse this skill when the task is about a named remote machine and Plexus has amatching `REMOTE.md` profile. Profiles live in the Plexus data directory:`$PLEXUS_DATA_DIR`, `$CLAUDE_PLUGIN_DATA`, `$CODEX_PLUGIN_DATA`, or `~/.plexus`as a fallback.## Required First StepIdentify the host from the user's request and run the context loader beforemaking changes:```bashpython3 "${CLAUDE_PLUGIN_ROOT:-${CODEX_PLUGIN_ROOT:-plugins/plexus}}/scripts/remote-context.py" <host>```Use `--no-probe` only when the user asks for an offline plan or when SSH/livetools are unavailable. Treat `<host>` as a profile name, not a shell fragment;use only the exact host token from the request or from known Plexus profiles.## How To Use The ContextTreat `REMOTE.md` as durable host memory: roles, important paths, accesspatterns, guardrails, and known quirks. Treat live probe output as the currentstate. If the two disagree, call out the discrepancy and prefer observed livestate for operational decisions.## Operating Rules- Use non-interactive SSH: `ssh -o BatchMode=yes <host> <command>`.- Inspect before changing state: uptime, disk, memory, failed units, Docker containers, and recent syslog entries.- Prefer host-specific guardrails in `REMOTE.md` over generic assumptions.- For service work, capture enough before/after evidence to prove the change.- If a task touches a reverse proxy, certificate, firewall, storage pool, or auth boundary, validate first and use a rollback path.## Related ContextPlexus pairs well with `cortex:cortex-logs`: recent logs and AI/session historyexplain what changed before the current request. It also composes withservice-specific skills like `tailscale`, `unraid`, `create-swag-config`, and`homelab-map`.

Frequently asked questions