CtrlK
BlogDocsLog inGet started
Tessl Logo

chrome-automation

Connect to and control Google Chrome browser using agent-browser with CDP (Chrome DevTools Protocol). Use when the user wants to automate their existing Chrome browser, see browser actions in real-time, or needs to control the Chrome instance they're already using. Handles installation, setup, connecting via remote debugging, and all browser automation tasks with live visual feedback.

67

Quality

82%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Critical

Do not install without reviewing

SKILL.md
Quality
Evals
Security

Quality

Content

65%

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

The body is highly actionable with concrete executable commands and real supporting scripts, but it is heavily redundant — repeating the install and Rust-fix workflows multiple times — and embeds detail that should live in the existing reference files. Workflow steps are clear but lack explicit validation feedback loops for the failure-prone build operations.

Suggestions

Deduplicate the install and Rust troubleshooting content: keep one canonical copy in the body and route the rest to references/troubleshooting.md instead of restating the same PATH/cargo fixes four times.

Replace the hard-coded absolute path '/Users/jyxc-dz-0100272/Documents/agent-browser' with a variable (e.g., $AB_HOME) and document it once, so commands stay copy-paste ready without the bloat.

Add an explicit validate→fix→retry checkpoint after the native binary build (e.g., 'Run agent-browser --version; on failure re-run npm run build:native with ~/.cargo/bin in PATH') to give the install workflow a real feedback loop.

DimensionReasoningScore

Conciseness

The body is ~670 lines with heavy repetition — the auto-install workflow and Rust PATH/cargo fixes are restated four times across the critical section, Quick Start, Installation, and Troubleshooting — and every command is padded with a hard-coded absolute path ('/Users/jyxc-dz-0100272/Documents/agent-browser'). It is mostly commands rather than concept explanations, but it could be tightened dramatically.

2 / 3

Actionability

Concrete, executable, copy-paste-ready commands throughout (open, snapshot -i, fill, click, screenshot) backed by real scripts in ./scripts/ that the body references, with platform-specific Mac and Windows variants.

3 / 3

Workflow Clarity

Steps are clearly numbered (auto-install steps, Core Workflow 1-6, real-time feedback protocol), but the build/install operations that can fail lack an explicit validate→fix→retry feedback loop within the main flow — 'verify' appears only as a final script step rather than a checkpoint.

2 / 3

Progressive Disclosure

One-level-deep references to references/setup-mac.md, setup-windows.md, and troubleshooting.md are present and the files exist, but the body keeps large amounts of detail inline (e.g., the entire Rust troubleshooting walkthrough is duplicated inline instead of deferred to the troubleshooting reference), making it more monolithic than an overview.

2 / 3

Total

9

/

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.

The description is strong: it states concrete capabilities, includes natural trigger terms a user would actually say, and clearly covers both what the skill does and when to use it. It is specific enough to avoid conflicting with other browser-oriented skills.

DimensionReasoningScore

Specificity

Names multiple concrete actions — 'Connect to and control Google Chrome browser', 'Handles installation, setup, connecting via remote debugging, and all browser automation tasks with live visual feedback' — rather than vague language.

3 / 3

Completeness

Explicitly answers both what (connect/control Chrome via CDP, installation, automation) and when ('Use when the user wants to automate their existing Chrome browser, see browser actions in real-time, or needs to control the Chrome instance').

3 / 3

Trigger Term Quality

Includes natural phrases users would say such as 'automate their existing Chrome browser', 'see browser actions in real-time', and 'control the Chrome instance they're already using'.

3 / 3

Distinctiveness Conflict Risk

The niche is specific — controlling the user's existing Chrome via CDP with real-time visual feedback — and clearly distinct from generic browser-automation skills, with explicit triggers unlikely to fire for the wrong skill.

3 / 3

Total

12

/

12

Passed

Validation

87%

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

Validation14 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

skill_md_line_count

SKILL.md is long (671 lines); consider splitting into references/ and linking

Warning

referenced_paths_exist

Referenced path issues: 4 missing

Warning

Total

14

/

16

Passed

Repository
anbeime/skill
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.