debug-node

Use when a Hetzner node is unreachable, SSH fails, cloud-init seems broken, or provisioning hangs. Boots into rescue mode via hcloud CLI to inspect filesystem, logs, SSH keys, sshd config, and cloud-init state without needing SSH access to the node itself.

Quality

89%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

78%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A highly actionable, well-sequenced rescue-debugging runbook with executable commands and a useful diagnosis table. Its main weaknesses are a monolithic single-file structure with dangling references to non-existent bundle files and the absence of an explicit validation loop around destructive fixes.

Suggestions

Create the referenced bundle files (docs/selinux.md, scripts/destroy.sh, scripts/cleanup.sh) or remove the references, so navigation targets actually resolve.

Split the SELinux guidance and destroy/cleanup procedures into references/ files and link to them from SKILL.md, keeping the main file as a lean overview-plus-steps.

Add an explicit validation checkpoint after Step 6 fixes (e.g., re-read /etc/shadow and authorized_keys, then re-run the Step 7 ssh probe with a 'if it fails, return to Step 5' loop) to satisfy the destructive-operation feedback-loop expectation.

Dimension	Reasoning	Score
Conciseness	Mostly lean and command-driven with minimal fluff, though the Overview reprises the description and the Leap Micro mental-model table, while justified domain knowledge, adds prose a tighter version could trim.	4 / 5
Actionability	Copy-paste-ready commands throughout (hcloud, mount, journalctl, sed, chroot ssh-keygen) with clear placeholders, plus a Common Diagnoses table covering the typical cases.	5 / 5
Workflow Clarity	Steps 1–7 are clearly sequenced with a structured diagnostic checklist and a final 'ssh ... echo ok' verification, but destructive fixes (shadow edits, host-key regen, reboot) lack an explicit validate-then-retry loop before proceeding.	4 / 5
Progressive Disclosure	Well-sectioned single file, but it is a ~285-line monolithic runbook and references docs/selinux.md, scripts/destroy.sh, and scripts/cleanup.sh — none of which exist in the bundle — so references are unsignaled/missing and inline content that could be split out remains inlined.	3 / 5
	Total	16 / 20 Passed

Description

100%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

An exemplary description: third-person voice, explicit 'Use when' triggers tied to natural symptoms, concrete inspection actions, and a distinct Hetzner-rescue niche. It matches the rubric's good examples closely with no padding or over-claims.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete inspection actions — 'inspect filesystem, logs, SSH keys, sshd config, and cloud-init state' via 'rescue mode' and 'hcloud CLI' — giving comprehensive coverage rather than vague claims.	5 / 5
Completeness	Explicit 'Use when...' clause with concrete triggers answers 'when', and 'Boots into rescue mode...to inspect...' answers 'what' with specific actions, satisfying both requirements.	5 / 5
Trigger Term Quality	Natural user-facing phrases — 'node is unreachable, SSH fails, cloud-init seems broken, or provisioning hangs' — are exactly what someone would say, with several symptom synonyms covered.	5 / 5
Distinctiveness Conflict Risk	Tightly scoped to 'Hetzner node' rescue debugging via 'hcloud CLI', a clear niche with distinct triggers and minimal overlap risk with other skills.	5 / 5
	Total	20 / 20 Passed

Validation

93%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
referenced_paths_exist	Referenced path issues: 2 missing	Warning

	Total	15 / 16 Passed

Repository: mysticaltech/terraform-hcloud-kube-hetzner
Path: .claude/skills/debug-node/SKILL.md
Commit: bb1622c

Reviewed: 1 day ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.