browser-guide

Best practices for using the managed browser — handling login walls, CAPTCHAs, lazy-loaded content, paywalls, and tab cleanup. Target=camoufox 主力路径。

Quality

80%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

High

Do not use without reviewing

Fix and improve this skill with Tessl

tessl review fix ./skills/browser-guide/SKILL.md

Quality

Content

77%

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 clear, validated workflows, but it is a verbose monolith that repeats the headed/headless rules and buries actionable detail under anecdotal padding. Splitting reference material into bundle files would meaningfully improve it.

Suggestions

Consolidate the headed/headless rules, which are stated in §0.1 and restated in §1-B, into one canonical block and cross-reference it to remove ~20 lines of redundancy and raise conciseness.

Move per-platform login specifics and the eval Correct/Wrong examples into reference files (e.g. references/login-flows.md, references/eval-patterns.md) linked from SKILL.md to enable proper progressive disclosure.

Trim anecdotal padding such as "真实事故：72 open / 1 close" and "critical — 机器会被撑死" down to one-line rules so every token earns its place.

Dimension	Reasoning	Score
Conciseness	Content is tool-specific and actionable rather than explaining generic concepts, but it is padded: the headed/headless rules appear in both §0.1 and §1-B, and anecdotal prose ("真实事故：72 open / 1 close", "critical — 机器会被撑死") could be tightened to one-line rules, so the score-2 anchor fits.	2 / 3
Actionability	Fully executable guidance throughout: concrete camoufox-cli invocations (e.g. "camoufox-cli --session <s> --json screenshot /tmp/qr-<platform>.png"), specific timeouts (3s/3min/5min), retry caps (2), and copy-paste eval examples with explicit Correct/Wrong blocks — matching the score-3 anchor.	3 / 3
Workflow Clarity	Multi-step processes (login flows 1-A–1-E, CAPTCHA fallback, lazy-load) are explicitly sequenced with validation checkpoints (snapshot to verify content loaded, poll QR/URL), retry limits, timeouts, and a batch teardown ("close --all") as an error-recovery loop, matching the score-3 anchor.	3 / 3
Progressive Disclosure	No bundle files exist and the entire ~200-line skill is a single monolithic SKILL.md with no one-level-deep references; content that could be split (per-platform login rules, eval patterns) is inline, matching the score-2 anchor of structure present but content that should be separate kept inline.	2 / 3
	Total	10 / 12 Passed

Description

82%

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 specific and trigger-rich on the capability side but lacks an explicit use-when clause, capping completeness. The trailing implementation jargon is out of place in a description.

Suggestions

Add an explicit trigger clause, e.g. ". Use when driving a browser against pages that show login walls, CAPTCHAs, paywalls, or lazy-loaded content." to raise completeness from 2 to 3.

Drop the implementation jargon "Target=camoufox 主力路径" from the description; it is not a user-facing trigger term and belongs in the body, not the frontmatter.

Keep the concrete action list but rephrase "Best practices for using the managed browser" into a third-person capability verb phrase (e.g. "Drives a managed browser to handle...") for tighter specificity.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — "handling login walls, CAPTCHAs, lazy-loaded content, paywalls, and tab cleanup" — matching the score-3 anchor of enumerating specific concrete actions rather than vague language.	3 / 3
Completeness	It states what the skill does but never provides a "Use when..." clause or equivalent explicit trigger guidance, so per the guidelines completeness is capped at 2 even though the what-side is strong.	2 / 3
Trigger Term Quality	Natural user-facing terms dominate ("login walls", "CAPTCHAs", "paywalls", "lazy-loaded content", "tab cleanup") that a user would actually say when needing this skill; only the trailing "Target=camoufox 主力路径" fragment is jargon.	3 / 3
Distinctiveness Conflict Risk	The combination of "managed browser" plus the specific wall/problem types carves a clear niche unlikely to fire for unrelated skills; the camoufox binding further distinguishes it.	3 / 3
	Total	11 / 12 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
metadata_version	'metadata.version' is missing	Warning
metadata_field	'metadata' should map string keys to string values	Warning

	Total	14 / 16 Passed

Repository: TeamWiseFlow/xiaobei
Path: skills/browser-guide/SKILL.md
Commit: bcef823

Reviewed: 3 days 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.