CtrlK
BlogDocsLog inGet started
Tessl Logo

browser-automation

统一自动化入口。覆盖浏览器自动化(Playwright)和 Windows 桌面应用自动化(OpenReverse)。 浏览器场景:打开网页、点击、填表、爬取、截图、自动化登录、渗透页面交互。 桌面场景:操作 IDA/x64dbg 等 GUI 工具、Windows UI Automation、视觉驱动交互、桌面应用网络抓包。 触发关键词:浏览器自动化、桌面自动化、打开网页、填表、爬取、截图、自动化登录、Playwright、agent-browser、headless、OpenReverse、UIA、CUA、桌面操作、Windows 自动化。

64

Quality

76%

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

Fix and improve this skill with Tessl

tessl review fix ./skills/browser-automation/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

65%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.

Highly actionable content with executable commands and clear routing tables, weakened by duplicated install instructions, missing validation feedback loops for batch operations, and an inlined command reference that overlaps with an unreferenced cheatsheet file.

Suggestions

Add explicit validation checkpoints to the browser and batch workflows (e.g. after a form submit, verify success before proceeding; on failure, retry), to lift workflow_clarity above the batch-operation cap.

De-duplicate the OpenReverse install steps — keep them in one place and reference that block from the bootstrap-prompt template instead of repeating the commands.

Move the full agent-browser command reference into references/playwright-cheatsheet.md and link to it from the body, keeping only the core 4-step workflow inline so the reference file is actually signaled and used.

DimensionReasoningScore

Conciseness

Mostly efficient with dense tables and code blocks, but the OpenReverse install steps are duplicated nearly verbatim in both '安装与配置' and 'OpenReverse 手动安装引导', and the agent-browser commands repeat between the workflow and command-reference sections — could be tightened.

3 / 5

Actionability

Provides copy-paste-ready, executable commands throughout — `agent-browser open <url>`, `agent-browser snapshot -i`, `agent-browser click @e1`, plus concrete git/npm install sequences — covering the common browser and install cases fully.

5 / 5

Workflow Clarity

The core workflow is sequenced (open → snapshot → operate → close) with a final self-check checklist, but batch/automated operations (批量表单提交, batch IDA analysis) lack inline validation feedback loops (validate → fix → retry), which caps workflow clarity at 3 per the batch-operations rule.

3 / 5

Progressive Disclosure

Section structure is present (Part 1/Part 2/bootstrap/routing), but an inlined agent-browser command reference overlaps with the existing references/playwright-cheatsheet.md, and that reference file is never linked from the body — content that belongs in a separate file is inlined and the reference is not signaled.

3 / 5

Total

14

/

20

Passed

Description

87%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, specific description with comprehensive trigger-term coverage and concrete actions spanning both browser and desktop automation. Its main weakness is conveying 'when' as a keyword list rather than a scenario-oriented 'Use when...' clause, leaving a small overlap risk with neighboring skills.

Suggestions

Replace or augment the '触发关键词' keyword list with an explicit scenario-framed clause such as 'Use when the user needs to automate web pages or Windows desktop GUIs...' to make the 'when' more explicit.

Add a one-line boundary statement clarifying when to prefer this skill over js-reverse or ida-reverse to reduce overlap/conflict risk.

Consider trimming the keyword dump to the highest-signal terms; comprehensive coverage is good but a few entries (e.g. 'headless') add limited trigger value.

DimensionReasoningScore

Specificity

Lists multiple concrete actions across both domains — '打开网页、点击、填表、爬取、截图、自动化登录、渗透页面交互' and '操作 IDA/x64dbg 等 GUI 工具、Windows UI Automation、视觉驱动交互、桌面应用网络抓包' — giving comprehensive, specific coverage.

5 / 5

Completeness

Clearly states 'what' (unified browser + desktop automation entry) and provides 'when' via an explicit trigger-keyword list, but the 'when' is a keyword dump rather than a scenario-framed 'Use when...' clause, so it could be more explicit.

4 / 5

Trigger Term Quality

The '触发关键词' line provides comprehensive natural-term coverage including synonyms and tool names: '打开网页、填表、爬取、截图、自动化登录、Playwright、agent-browser、headless、OpenReverse、UIA、CUA'.

5 / 5

Distinctiveness Conflict Risk

The Playwright/OpenReverse split with distinct tool-name triggers carves a clear niche, but combining browser and desktop automation under one '统一自动化入口' creates minor overlap risk with related specialized modules (js-reverse, ida-reverse).

4 / 5

Total

18

/

20

Passed

Validation

100%

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

Validation16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
zhaoxuya520/reverse-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.