agentleFS
understudy-aiunderstudy-ai/understudy/skills/healthcheck/SKILL.md

healthcheck

Skill461 starsChanged 8 months agoMITView on GitHub

Host security hardening and risk-tolerance guidance for Understudy deployments. Use when a user asks for security audits, firewall/SSH/update hardening, exposure review, or recurring checks on a machine running Understudy.

38 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use taskflow-inbox-triage, node-connect and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • AtomicBot-aiSkill

    taskflow-inbox-triage

    Also in 6 projects that use this

  • hkyutongSkill

    node-connect

    Diagnose OpenClaw node connection and pairing failures for Android, iOS, and macOS companion apps. Use when QR/setup code/manual connect fails, local Wi-Fi works but VPS/tailnet does not, or errors mention pairing required, unauthorized, bootstrap token invalid or expired, gateway.bind, gateway.remote.url, Tailscale, or plugins.entries.device-pair.config.publicUrl.

    Also in 18 projects that use this

  • molt-botSkill

    discord-e2e

    Test Discord with Convex-leased QA bots: check credentials without a Gateway, exercise owned messages, or prove OpenClaw replies, files, threads, reactions, streaming, and typing. Distinguish API and Gateway evidence from manual user interactions.

    Also in 10 projects that use this

  • molt-botSkill

    openclaw-repair-sweep

    Run scoped OpenClaw issue/PR repair campaigns: coordinate workers, prove root causes, and land or close verified work under the requested authority.

    Also in 10 projects that use this

  • clawdbotSkill

    openclaw-performance

    Performance work in OpenClaw: a slow or heavy flow (latency, CPU, memory, SQL, RPC, renders), writing or reviewing a perf PR, or adding a cache. Count the work, delete waste first, cache last.

    Also in 7 projects that use this

  • clawdbotSkill

    plugin-sdk-impact

    Count Plugin SDK consumers across live ClawHub releases and bundled OpenClaw plugins before deprecation, removal, or behavior changes; supports public release-date filters.

    Also in 7 projects that use this

  • +2 more from molt-bot

The file

Show its rules and text

We split this skill into the rules it gives an agent

21 rules. 0 are common, 16 are distinctive, and 5 are very specific to this project.

  • 0

    Common

    Most of the 37 edited versions we read keep these.

  • 16

    Distinctive

    This author's own choices. Worth a look.

  • 5

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Start with read-only checks.

    line 15
  2. Distinctive:

    Do not modify remote access settings until the current access path is confirmed.

    line 17
  3. Distinctive:

    Prefer reversible changes with a rollback plan.

    line 18
  4. Very specific:

    Never claim Understudy itself changes the host firewall, SSH policy, or OS update settings.

    line 19
  5. Distinctive:

    If identity or role is unclear, provide recommendations only.

    line 20
  6. Distinctive:

    Ask once for permission to run read-only checks.

    line 43
  7. Distinctive:

    Always present the plan before making changes.

    line 56
  8. Distinctive:

    stop on unexpected output

    line 72
  9. Distinctive:

    After any change, re-check:

    line 87
  10. Very specific:

    Use Understudy's canonical scheduling surfaces:

    line 98
  11. Distinctive:

    Prefer stable job names:

    line 102
  12. Very specific:

    Before creating or updating, inspect existing jobs with schedule action "list" or understudy schedule --list.

    line 105
  13. Distinctive:

    Do not create recurring jobs without explicit approval.

    line 110
  14. Distinctive:

    Use only commands that exist in this repo:

    line 114
  15. Very specific:

    Do not invent understudy update status, understudy cron ..., unsupported schedule subcommands, or flags that are not in this repo.

    line 122
  16. Distinctive:

    If the user wants an audit trail, record:

    line 126
  17. Distinctive:

    Never log tokens or full credential contents.

    line 133
  18. Distinctive:

    Only write to memory files when the user explicitly opts in and the session is private or local.

    line 137
  19. Distinctive:

    If memory is requested:

    line 138
  20. Very specific:

    update MEMORY.md only for durable preferences such as risk posture or allowed ports

    line 141
  21. Distinctive:

    If the session cannot write to the workspace, provide a redacted summary the user can store manually.

    line 144

“Common” is measured: kept by at least 6 in 10 of the 37 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Understudy Host Hardening

Overview

Assess the host that runs Understudy, keep host controls separate from Understudy configuration, and require explicit approval before any change.

Core Rules

  • Start with read-only checks.
  • Require explicit approval before any state-changing action.
  • Do not modify remote access settings until the current access path is confirmed.
  • Prefer reversible changes with a rollback plan.
  • Never claim Understudy itself changes the host firewall, SSH policy, or OS update settings.
  • If identity or role is unclear, provide recommendations only.
  • Present user choices as numbered lists.

Read-Only Workflow

  1. Establish context:
    • OS and version
    • local console vs SSH/RDP/tunnel/tailnet access
    • admin/root availability
    • public exposure, reverse proxy, or private-only reachability
    • backups, disk encryption, and automatic security updates
  2. Run host checks:
    • uname -a, sw_vers, cat /etc/os-release
    • Linux: ss -ltnup, ufw status, firewall-cmd --state, nft list ruleset
    • macOS: lsof -nP -iTCP -sTCP:LISTEN, /usr/libexec/ApplicationFirewall/socketfilterfw --getglobalstate, pfctl -s info
    • macOS backups: tmutil status
  3. Run Understudy checks:
    • understudy security --audit
    • understudy status --json
    • understudy health --json
    • understudy logs --tail 100
    • understudy doctor --deep when broader local diagnostics are needed

Ask once for permission to run read-only checks. If granted, infer as much as possible from commands before asking follow-up questions.

Risk Profiles

Offer one of these numbered defaults after the host context is known:

  1. Home / workstation balanced
  2. Remote / headless hardened
  3. Developer convenience with explicit exposure warnings
  4. Custom constraints supplied by the user

Remediation Plan

Always present the plan before making changes. Include:

  • target profile
  • current posture summary
  • gaps vs target
  • exact commands to run
  • rollback or access-preservation notes
  • credential and file-permission hygiene notes

Execution Rules

For each state-changing step:

  • show the exact command first
  • explain impact and rollback
  • confirm remote access will remain available
  • stop on unexpected output

Require explicit approval for:

  • firewall changes
  • opening or closing ports
  • SSH or RDP configuration changes
  • package installation or removal
  • enabling or disabling services
  • user or group changes
  • scheduled jobs or persistence
  • access to sensitive files or credentials

Verification

After any change, re-check:

  • firewall status
  • listening ports
  • remote access still works
  • understudy security --audit

Deliver a final posture summary and note any deferred items.

Periodic Checks

Use Understudy's canonical scheduling surfaces:

  • In agent/runtime workflows: use the schedule tool with actions status, create, list, update, remove, run, runs
  • For explicit manual CLI instructions: use the real understudy schedule command with supported flags such as --status, --list, --add, --update, --remove, --run, and --runs
  • Prefer stable job names:
    • healthcheck:security-audit
    • healthcheck:status-snapshot
  • Before creating or updating, inspect existing jobs with schedule action "list" or understudy schedule --list.
  • Example recurring commands:
    • weekly understudy security --audit
    • daily understudy status --json

Do not create recurring jobs without explicit approval.

Command Accuracy

Use only commands that exist in this repo:

  • understudy security --audit
  • understudy status --json
  • understudy health --json
  • understudy logs --tail 100
  • understudy doctor --deep

Do not invent understudy update status, understudy cron ..., unsupported schedule subcommands, or flags that are not in this repo.

Recordkeeping

If the user wants an audit trail, record:

  • plan ID and timestamp
  • approved steps and exact commands
  • exit codes and files modified (best effort)
  • redacted findings only

Never log tokens or full credential contents.

Memory Writes (Conditional)

Only write to memory files when the user explicitly opts in and the session is private or local. If memory is requested:

  • append dated notes to memory/YYYY-MM-DD.md
  • update MEMORY.md only for durable preferences such as risk posture or allowed ports
  • redact hostnames, IPs, usernames, serials, service names, and secrets

If the session cannot write to the workspace, provide a redacted summary the user can store manually.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.