CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

claude-in-chrome-troubleshooting

Diagnose and fix Claude in Chrome MCP extension connectivity issues. Use when mcp__claude-in-chrome__* tools fail, return "Browser extension is not connected", or behave erratically.

Quality

80%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./skills/claude-in-chrome-troubleshooting/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

71%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 troubleshooting skill with executable commands, a complete toggle script, and a verification-checked reset procedure. Its weaknesses are redundancy between the two diagnosis sections and a monolithic single-file structure where a reference file would reduce context load.

Suggestions

Merge "Quick Diagnosis" and "Diagnostic Deep Dive" into one section (they repeat the same ps/socket/config/TMPDIR checks) and drop the Summary section, which restates the body — this would save roughly 40 lines of context.

Move the "Diagnostic Deep Dive" commands and the File Reference table into a references/ file (e.g., DIAGNOSTICS.md) linked from a short pointer, keeping SKILL.md as a lean overview of the primary fix and reset procedure.

Add a short failure-recovery loop after reset steps 6-7: what to check if the native host is not running or the socket does not appear (e.g., re-check active configs, confirm the extension icon was clicked), closing the workflow's missing feedback loop.

DimensionReasoningScore

Conciseness

Commands are tight and the content is largely novel (socket formats, config conflicts Claude would not know), but "Quick Diagnosis" and "Diagnostic Deep Dive" substantially duplicate each other (both run ps aux, check both socket locations, list configs, check TMPDIR), and the Summary section restates prior content. This fits 'mostly efficient but could be tightened' rather than the trim-every-token 4/5 anchors.

3 / 5

Actionability

Fully executable, copy-paste-ready bash throughout: the config-disable mv commands, a complete chrome-mcp-toggle shell function, and an 8-step reset procedure including a heredoc wrapper script. Specific examples cover the common cases (CLI vs Cowork, stale wrapper, TMPDIR unset), matching the 5 anchor.

5 / 5

Workflow Clarity

The Full Reset Procedure is a clearly numbered 8-step sequence with explicit verification checkpoints (step 6 'Verify correct native host is running', step 7 'Verify socket exists'), and the "Critical Insight" section explains the restart ordering. It falls short of the 5 anchor because there is no explicit feedback loop — no guidance on what to do if verification at steps 6/7 fails — placing it at 'clear sequence with most checkpoints present'.

4 / 5

Progressive Disclosure

The single ~245-line file is well-sectioned with clear headers, but everything is inlined: the Diagnostic Deep Dive and File Reference tables are exactly the kind of material that belongs in a separate reference file per the rubric's guidance, and the skill exceeds the under-50-lines threshold where inline-only structure can score 5. It sits at 'some structure but content that should be separate is inline' rather than the well-split 4 anchor.

3 / 5

Total

15

/

20

Passed

Description

82%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.

A strong description: explicit what-and-when structure with highly specific, literal trigger terms drawn from the actual failure mode. The only gap is that the capability statement stays at the generic 'diagnose and fix' level rather than naming the concrete fixes it covers.

DimensionReasoningScore

Specificity

The description names the domain ("Claude in Chrome MCP extension connectivity issues") and two concrete actions ("Diagnose and fix"), but does not break capabilities into finer-grained actions like the multi-action 4/5 anchors. It fits the 'names domain and 1-2 concrete actions' anchor; the 4 anchor expects several specific actions, which this lacks.

3 / 5

Completeness

It explicitly answers what ("Diagnose and fix Claude in Chrome MCP extension connectivity issues") and when ("Use when mcp__claude-in-chrome__* tools fail, return 'Browser extension is not connected', or behave erratically") with concrete trigger phrases, matching the 5 anchor exactly. The 4 anchor requires a 'when' that could be more explicit; here the triggers are specific and literal.

5 / 5

Trigger Term Quality

Triggers include the literal tool pattern ("mcp__claude-in-chrome__* tools fail"), the exact error string ("Browser extension is not connected"), and "behave erratically" — natural terms a user or Claude would actually encounter. A few natural variations (e.g., "browser automation", "Chrome extension not working") are missing, keeping it below the comprehensive 5 anchor.

4 / 5

Distinctiveness Conflict Risk

A clear niche — the Claude in Chrome MCP extension specifically — with distinct trigger terms (tool prefix pattern, exact error message) that no general Chrome-automation or connectivity skill would match. Minimal conflict risk per the 5 anchor.

5 / 5

Total

17

/

20

Passed

Validation

93%

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

CriteriaDescriptionResult

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

15

/

16

Passed

Repository

administrakt0r/AI-Agents-Safe-Coding-Skills

Path

skills/claude-in-chrome-troubleshooting/SKILL.md

Reviewed

Table of Contents

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.