CtrlK
BlogDocsLog inGet started
Tessl Logo

browser-ops-skill

当用户要求对具体网页做页面级交互时触发:登录、填表、点按钮、翻页抓取、截图、下载页面图片、处理验证码或反爬、访问 SPA 动态内容;也用于 WebFetch 拿到空壳或被风控后的降级。典型说法"帮我登录 X 然后抓 Y"、"截个图"。**不用于**:事实性问题或"xx 是什么"(→`WebSearch`)、单纯读取静态网页文本(先试 `WebFetch`)。

75

Quality

94%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

Quality

Content

88%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 lean, highly actionable body built around an explicit observe-act-verify loop with concrete tool calls, validation checkpoints, and feedback loops. Its only soft spots are mild redundancy in the 铁律 summary and the absence of any external reference structure.

Suggestions

Trim the 铁律 section or fold it into the relevant workflows, since several rules ('先 snapshot 再行动', '每次操作后 BrowserSnapshot') already appear earlier—this would recover tokens without losing guidance.

Consider moving the per-tool detail (BrowserSnapshot return fields, wait-strategy table) into a references/ file referenced once from the body, improving progressive_disclosure toward a one-level-deep structure.

Replace the soft commentary lines ('截图是调试利器:文本看不出问题时,看截图一目了然') with the concrete instruction they imply, to keep every line actionable.

DimensionReasoningScore

Conciseness

Dense and tactical throughout—tool tables, selector priority, fill-vs-type, wait-strategy table—assuming Claude's competence; the '铁律' list largely restates points already made (e.g. '先 snapshot 再行动'), a minor redundancy that could be trimmed.

4 / 5

Actionability

Concrete tool invocations with parameters appear throughout (e.g. BrowserAct { action: 'fill', selector: '...', text: '...' }, BrowserWait { type: 'urlIncludes', url: '/success' }), covering the three named workflows in copy-paste-ready form.

5 / 5

Workflow Clarity

Three named workflows with explicit sequences plus the observe-act-verify loop enforcing a snapshot checkpoint after every action; the error-diagnosis section and 铁律 provide feedback loops and a checklist, including per-page validation for the paginated batch workflow.

5 / 5

Progressive Disclosure

No bundle files exist and all content is inline in one ~110-line file with clear section headers and no nested references; well-organized and self-contained, though slightly above the simple-skill threshold where one-level references would be expected for a 5.

4 / 5

Total

18

/

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.

A tightly written, third-person description that names concrete capabilities, gives natural user trigger phrases, and draws clear boundaries against WebSearch/WebFetch. It hits every dimension at the top anchor.

DimensionReasoningScore

Specificity

Lists multiple concrete page-level actions—'登录、填表、点按钮、翻页抓取、截图、下载页面图片、处理验证码或反爬、访问 SPA 动态内容'—with comprehensive coverage and an added degradation use case, matching the score-5 anchor.

5 / 5

Completeness

Explicitly answers both what (page-level interactions) and when ('当用户要求对具体网页做页面级交互时触发' plus '典型说法' examples), with concrete trigger phrases and 不用于 exclusions.

5 / 5

Trigger Term Quality

Quotes natural user utterances directly ('帮我登录 X 然后抓 Y', '截个图') and contrasts with sibling tools WebSearch/WebFetch, giving comprehensive natural-term coverage including synonyms.

5 / 5

Distinctiveness Conflict Risk

The explicit '不用于' boundary (factual questions → WebSearch, static text → WebFetch) carves a clear niche with minimal conflict risk against sibling skills.

5 / 5

Total

20

/

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.

Validation15 / 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
itsablabla/openloaf-web
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.