tunnel-doctor

Diagnoses and fixes conflicts between Tailscale and proxy/VPN tools (Shadowrocket, Clash, Surge) on macOS. Covers five conflict layers - (1) route hijacking, (2) HTTP proxy env var interception, (3) system proxy bypass, (4) SSH ProxyCommand double tunneling, and (5) VM/container runtime proxy propagation (OrbStack/Docker). Includes SOP for remote development via SSH tunnels with proxy-safe Makefile patterns. Use when Tailscale ping works but SSH/HTTP times out, when browser returns 503 but curl works, when git push fails with "failed to begin relaying via HTTP", when Docker pull times out behind TUN/VPN, when setting up Tailscale SSH to WSL instances, or when bootstrapping remote dev environments over Tailscale.

Quality

92%

Does it follow best practices?

Impact

—

No eval scenarios have been run

Securityby

Advisory

Suggest reviewing before use

Quality

Content

85%

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

This is a high-quality, expert-level skill document that excels at actionability and workflow clarity. The five-layer conflict model provides an excellent mental framework, and the symptom-based diagnostic approach is highly practical. The main weakness is moderate verbosity—some explanations of 'why' could be trimmed since Claude can infer rationale from the technical context, and there's some repetition between the diagnostic and SOP sections.

Suggestions

Consolidate repeated 'why this works' explanations—the technical context makes most rationale self-evident to Claude

Consider moving the detailed sub-problem explanations in Step 2G to the reference file, keeping only the diagnostic commands and fix commands in the main skill

Dimension	Reasoning	Score
Conciseness	The skill is comprehensive but includes some redundant explanations (e.g., repeated explanations of why certain approaches work/don't work across sections). The tables and diagnostic workflows are efficient, but some sections like 2G could be tightened. The SOP section repeats concepts already covered in the diagnostic workflow.	2 / 3
Actionability	Excellent actionability with fully executable commands, specific bash/makefile code examples, and copy-paste ready configurations. Every diagnostic step includes concrete commands to run and expected output patterns to compare against. The verification steps are explicit and testable.	3 / 3
Workflow Clarity	Outstanding workflow structure with clear symptom-to-solution mapping in Step 1, explicit validation commands after each fix, and a comprehensive end-to-end verification in Step 6. The diagnostic workflow uses a decision tree approach that guides through the five conflict layers systematically with clear checkpoints.	3 / 3
Progressive Disclosure	Well-structured with a clear overview table of the five conflict layers, symptom-based navigation in Step 1, and appropriate delegation to references/proxy_conflict_reference.md for detailed per-tool configurations. The SOP section is appropriately separated from diagnostic content.	3 / 3
	Total	11 / 12 Passed

Description

100%

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

This is an exemplary skill description that excels across all dimensions. It provides highly specific technical capabilities organized into five conflict layers, includes natural trigger phrases that match real user problem descriptions, and has an explicit 'Use when...' clause with six distinct scenarios. The combination of specific tools and failure modes makes it highly distinctive.

Dimension	Reasoning	Score
Specificity	Lists multiple specific concrete actions across five distinct conflict layers (route hijacking, HTTP proxy env var interception, system proxy bypass, SSH ProxyCommand double tunneling, VM/container runtime proxy propagation) with named tools (Tailscale, Shadowrocket, Clash, Surge, OrbStack, Docker).	3 / 3
Completeness	Clearly answers both what (diagnoses/fixes conflicts between Tailscale and proxy/VPN tools across five layers) AND when (explicit 'Use when...' clause with six specific trigger scenarios covering common failure modes).	3 / 3
Trigger Term Quality	Excellent coverage of natural terms users would say: 'Tailscale ping works but SSH times out', 'browser returns 503 but curl works', 'git push fails', 'Docker pull times out', 'Tailscale SSH to WSL'. These are exact phrases users experiencing these issues would use.	3 / 3
Distinctiveness Conflict Risk	Highly distinctive niche combining Tailscale + proxy tool conflicts on macOS. The specific tool combinations (Shadowrocket, Clash, Surge) and failure scenarios (Tailscale ping works but SSH times out) create clear, unique triggers unlikely to conflict with general networking or VPN skills.	3 / 3
	Total	12 / 12 Passed

Validation

81%

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 — 9 / 11 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (640 lines); consider splitting into references/ and linking	Warning
allowed_tools_field	'allowed-tools' contains unusual tool name(s)	Warning

	Total	9 / 11 Passed

Repository: fernandezbaptiste/claude-code-skills
Commit: 4f0eae8

Reviewed: 4 months 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.