CtrlK
BlogDocsLog inGet started
Tessl Logo

smart-search

智能搜索路由器。根据用户意图选择最佳搜索源,构造 URL 并导航,获取内容。 仅在默认搜索手段效果不佳、或需要登录自媒体平台(抖音/小红书/微博等)搜索时启用本技能; 常规通用搜索请先用 onboard 配置的搜索供应商(火山/serper/Tavily 等)。

66

Quality

81%

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

77%

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

A highly actionable, well-sequenced routing skill with concrete commands and solid error-recovery feedback loops. Its weaknesses are overlapping routing tables that bloat the body and per-site reference files that are referenced but missing from the bundle.

Suggestions

Provide (or include in the bundle) the referenced 'sites/<platform>.md' files so progressive disclosure is actually navigable.

Consolidate the three overlapping routing tables (platform table, Per-Category Source Guides, '用户未指定平台' table) to reduce redundancy and tighten conciseness.

Move the large static reference tables (warmup URLs, keyword encoding) into a single appendix or a 'sites/' reference file to keep the overview lean.

DimensionReasoningScore

Conciseness

Content is dense platform-specific operational knowledge (no basic-concept over-explanation), but the platform table, Per-Category Source Guides, and '用户未指定平台' table overlap in routing guidance and could be tightened — 'mostly efficient but could be tightened'.

2 / 3

Actionability

Provides concrete, copy-paste-ready commands ('camoufox-cli --session "$SESSION" --json open "<url>"'), specific warmup URLs, exact encoding rules ('#' → '%23'), and numeric rate caps — fully executable guidance.

3 / 3

Workflow Clarity

Sequenced '使用流程' and '搜索后操作' steps with explicit validation checkpoints (snapshot confirmation, 30s timeout + re-snapshot feedback loop, login-wall → browser-guide handoff), matching the clear-sequence-with-feedback anchor.

3 / 3

Progressive Disclosure

Overview is well-organized with clearly signaled one-level references to 'sites/<platform>.md', but those files are absent from the bundle, so the disclosure is not actually realized and navigation dead-ends — better than poorly-signaled refs but cannot be a 3.

2 / 3

Total

10

/

12

Passed

Description

85%

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 specific, well-scoped description with an explicit when-to-enable trigger and clear distinctiveness from general search skills. Its main weakness is trigger-term coverage, which leans on platform names rather than the natural search phrases a user might say.

Suggestions

Add natural-language trigger terms users would actually say (e.g. '搜索/搜一下/查找' alongside the platform names) to broaden trigger-term coverage.

Consider listing a couple of representative query intents (e.g. '在抖音上搜…', '搜小红书') as concrete triggers to push trigger_term_quality to 3.

DimensionReasoningScore

Specificity

Names multiple concrete actions — '选择最佳搜索源' (select best search source), '构造 URL 并导航' (construct URL and navigate), '获取内容' (fetch content) — matching the 'lists multiple specific concrete actions' anchor.

3 / 3

Completeness

Clearly answers what it does (route/construct/navigate/fetch) and when to use it via an explicit '仅在…启用本技能' trigger clause, satisfying both 'what' and 'when' with explicit triggers.

3 / 3

Trigger Term Quality

Includes platform names users would say (抖音/小红书/微博) but misses common natural search-term variations like generic '搜索/查找/搜一下', so it has 'some relevant keywords but missing common variations'.

2 / 3

Distinctiveness Conflict Risk

Carves a distinct niche (login-required self-media platform search / fallback) and explicitly defers general search to onboard providers, making conflict with generic search skills unlikely.

3 / 3

Total

11

/

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

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
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.