CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

github-weekly-trending

解读用户粘贴的 GitHub 每周 Trending 或开源项目榜单内容,基于用户提供的候选项目、项目 README、GitHub 项目页和 DeepWiki 生成适合公众号发布的 Markdown 文章,并输出到当前项目 markdown 目录。用于 GitHub 周榜解读、每周开源热点、热门开源项目分析、公众号开源项目推荐。

72

Quality

91%

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%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.

The body is a highly actionable, well-sequenced workflow with strong validation gates (fact cards, submission checklist, failure handling), but it pays for that with token redundancy — the anti-scraping and no-fabrication rules each repeat 3–4 times — and keeps writing-style, chart, and comparison rulebooks inline in a single ~320-line file instead of splitting them into references.

Suggestions

State each guardrail once: consolidate the '不要主动抓取 GitHub Trending' rule (currently in Step 1 twice, the checklist, and the failure section) and the no-fabrication rule (currently in 数据来源优先级, Step 4, Step 5, and the checklist) into a single authoritative statement, leaving the checklist as the only reminder.

Move 写作风格, 图表规则, 竞品对比规则, and 代码示例规则 into reference files (e.g., references/writing-style.md, references/chart-rules.md) and link them from SKILL.md so the main file stays a lean workflow overview.

Trim redundant elaboration such as the banned-phrase variants list and the illustrative Mermaid example, which mostly restate rules already given in the 图表规则 section.

DimensionReasoningScore

Conciseness

Mostly efficient operational rules, but several are stated repeatedly: "不要自行抓取 GitHub Trending 页面替代用户输入" appears in Step 1 (twice: "不要主动访问 https://github.com/trending…" and "不要自行抓取 GitHub Trending 页面替代用户输入"), in the 质量校验 checklist, and in 失败和不确定情况; the no-fabrication rule ("禁止把未验证信息写成事实") recurs in 数据来源优先级, Step 4, Step 5, and the checklist. This is 'mostly efficient but could be tightened' (anchor 3), not anchor 4 — the repetition is systematic, not minor — and not anchor 2 since no section explains concepts Claude already knows.

3 / 5

Actionability

Fully concrete, copy-paste-ready guidance: README fallback URLs ("https://raw.githubusercontent.com/{owner}/{repo}/{default_branch}/README.md"), named DeepWiki tools (read_wiki_structure / read_wiki_contents / ask_question with repoName), a worked normalization example ("microsoft / markitdown" → repo URL → DeepWiki param), a fact-card field template, a full article markdown skeleton, and a sample Mermaid block. Matches anchor 5 for an instruction-only skill with actionable concrete guidance; not anchor 4 because the common cases (project selection, README retrieval, article structure, failure handling) are all covered with specific examples.

5 / 5

Workflow Clarity

The pipeline is diagrammed up front ("用户粘贴榜单内容 -> README / GitHub 项目页 -> DeepWiki -> 项目事实卡 -> 公众号文章 -> 当前项目 markdown 目录 -> 事实校验") and elaborated as six sequenced steps with an explicit validation checkpoint (Step 5 fact cards required before writing), a checkbox 质量校验 checklist before submission, and a 失败和不确定情况 section covering error recovery (missing fields, README-poor projects, DeepWiki gaps). This is anchor 5 (explicit validation steps, feedback loops, checklist); not anchor 4 since both checkpoints and recovery paths are present.

5 / 5

Progressive Disclosure

The ~320-line single-file body is well-sectioned, but content that reads as reference material is inlined: 写作风格, 图表规则 (Mermaid/SVG rules + example), 竞品对比规则, and 代码示例规则 could live in separate reference files, and no references/scripts/assets bundle exists. Anchor 3 fits ('some structure but could be better organized… content that should be separate is inline'); not anchor 4 because there are no external reference files at all despite the length, and not anchor 2 because section headers are clear and nothing is buried.

3 / 5

Total

16

/

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.

The description is exemplary: it states the full input-to-output pipeline in concrete terms (pasted trending list → README/GitHub/DeepWiki verification → Markdown article → project markdown directory) and closes with an explicit '用于…' clause enumerating natural trigger phrases. Both 'what' and 'when' are unambiguous and the niche is clearly distinct.

DimensionReasoningScore

Specificity

Quotes multiple specific concrete actions: "解读用户粘贴的 GitHub 每周 Trending 或开源项目榜单内容", "基于用户提供的候选项目、项目 README、GitHub 项目页和 DeepWiki 生成适合公众号发布的 Markdown 文章", "并输出到当前项目 markdown 目录". This matches the anchor-5 example's comprehensiveness (input source, verification inputs, output format, output location); it is not anchor 4 because there are no meaningful gaps in coverage of the skill's actions.

5 / 5

Completeness

Explicitly answers both what ("基于…README、GitHub 项目页和 DeepWiki 生成适合公众号发布的 Markdown 文章") and when ("用于 GitHub 周榜解读、每周开源热点、热门开源项目分析、公众号开源项目推荐") with concrete trigger phrases — the anchor-5 pattern. Not anchor 4 because the 'when' clause is explicit and enumerates multiple triggers rather than being merely adequate.

5 / 5

Trigger Term Quality

Natural trigger phrases users would say are comprehensively covered with synonyms: "GitHub 周榜解读、每周开源热点、热门开源项目分析、公众号开源项目推荐", plus "GitHub 每周 Trending" and "开源项目榜单" in the capability clause. Matches anchor 5 (comprehensive natural terms including variations); not anchor 4 since no common phrasing for this need is obviously missing.

5 / 5

Distinctiveness Conflict Risk

A clear niche — interpreting a user-pasted GitHub weekly trending list into a WeChat 公众号 article — with distinct triggers ("GitHub 每周 Trending", "周榜", "公众号开源项目推荐") unlikely to fire for other skills. Matches anchor 5; not anchor 4 since overlap risk with adjacent skills (generic GitHub analysis, generic writing) is minimal given the trigger wording.

5 / 5

Total

20

/

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.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository

digoal/blog

Path

skills/github-weekly-trending/SKILL.md

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.