How do I install the Background Supervisor Skill skill?

Run npx skillmds@latest add xukai-ye/background-supervisor-skill 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 Background Supervisor Skill safe to use?

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

Which AI agents work with Background Supervisor Skill?

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 Background Supervisor Skill free to use?

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

Who published Background Supervisor Skill?

Xukai-YE (@xukai-ye) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Background Supervisor Skill by xukai-ye

Name: Background Supervisor Skill
Rating: 4.666666666666667 (3 reviews)
Author: Xukai-YE

Background Supervisor Skill

Enable automatic Python-based supervision for long-running Claude Code background Bash tasks. Use when you want background programs to emit periodic monitor output and have Claude Code review, verify, and rerun jobs after failures. Installs and maintains hook-driven monitoring in ~/.claude/settings.json.

Xukai-YE Updated Sep 22, 2026

File contents

Background Supervisor

This skill turns long-running background Bash tasks into supervised jobs.

When active, the expected outcome is:

a Python monitor prints background status every 60 seconds
failed jobs are retried with bounded restart limits
exhausted restart budgets trigger a Claude Code review/repair attempt
verified repairs restart the original job

Use This Skill When

The user wants long-running background programs to be monitored automatically
The user wants periodic terminal output from background jobs
The user wants Claude Code to review and attempt repair when background jobs fail
The user wants the behavior wired into Claude Code globally through hooks

Workflow

Verify prerequisites:
- python is available on PATH
- claude CLI is available on PATH and authenticated
- The environment is Windows-oriented, because PID attachment uses tasklist and PowerShell process inspection
Install or refresh the global hooks:
- Run python "$CLAUDE_PLUGIN_ROOT/scripts/install.py"
Verify the install:
- Inspect ~/.claude/settings.json
- Confirm both PostToolUse and PostToolUseFailure have Bash hooks pointing to bg_hook.py
Explain the runtime model briefly:
- monitor interval is 60 seconds
- restart budget is 2
- repair budget is 1
- verified fixes are required before rerun
If the user wants to disable the behavior:
- Run python "$CLAUDE_PLUGIN_ROOT/scripts/uninstall.py"

Runtime Files

scripts/bg_hook.py: receives Claude hook events and persists job state
scripts/bg_monitor.py: watches job state, prints status, restarts jobs, and escalates repairs
scripts/bg_repair.py: invokes Claude Code non-interactively for bounded repair attempts
scripts/bg_common.py: shared helpers and recovery policy

Notes

This skill is intentionally narrow. It focuses on background Bash jobs, not general service orchestration.
It works best for commands that are clearly long-running and safe to restart.
For setup details, see references/setup.md.
For runtime behavior and safety limits, see references/runtime-model.md.

---name: background-supervisor-skilldescription: Enable automatic Python-based supervision for long-running Claude Code background Bash tasks. Use when you want background programs to emit periodic monitor output and have Claude Code review, verify, and rerun jobs after failures. Installs and maintains hook-driven monitoring in ~/.claude/settings.json.---# Background SupervisorThis skill turns long-running background Bash tasks into supervised jobs.When active, the expected outcome is:- a Python monitor prints background status every 60 seconds- failed jobs are retried with bounded restart limits- exhausted restart budgets trigger a Claude Code review/repair attempt- verified repairs restart the original job## Use This Skill When- The user wants long-running background programs to be monitored automatically- The user wants periodic terminal output from background jobs- The user wants Claude Code to review and attempt repair when background jobs fail- The user wants the behavior wired into Claude Code globally through hooks## Workflow1. Verify prerequisites: - `python` is available on PATH - `claude` CLI is available on PATH and authenticated - The environment is Windows-oriented, because PID attachment uses `tasklist` and PowerShell process inspection2. Install or refresh the global hooks: - Run `python "$CLAUDE_PLUGIN_ROOT/scripts/install.py"`3. Verify the install: - Inspect `~/.claude/settings.json` - Confirm both `PostToolUse` and `PostToolUseFailure` have Bash hooks pointing to `bg_hook.py`4. Explain the runtime model briefly: - monitor interval is 60 seconds - restart budget is 2 - repair budget is 1 - verified fixes are required before rerun5. If the user wants to disable the behavior: - Run `python "$CLAUDE_PLUGIN_ROOT/scripts/uninstall.py"`## Runtime Files- `scripts/bg_hook.py`: receives Claude hook events and persists job state- `scripts/bg_monitor.py`: watches job state, prints status, restarts jobs, and escalates repairs- `scripts/bg_repair.py`: invokes Claude Code non-interactively for bounded repair attempts- `scripts/bg_common.py`: shared helpers and recovery policy## Notes- This skill is intentionally narrow. It focuses on background Bash jobs, not general service orchestration.- It works best for commands that are clearly long-running and safe to restart.- For setup details, see `references/setup.md`.- For runtime behavior and safety limits, see `references/runtime-model.md`.

Frequently asked questions