BD NewsToday
NewsToday solves information overload for users who want to stay informed without spending an hour checking scattered sources. Instead of manually browsing Weibo, Zhihu, Baidu, and news apps separately, NewsToday aggregates, deduplicates, and summarizes the most important stories into a single readable briefing delivered at the right time. Every morning, NewsToday pushes a curated briefing of 10 top stories spanning politics, finance, technology, international affairs, and society — pulled from both RSS feeds (Sina News, The Paper, 36Kr, BBC Chinese, Reuters Chinese) and real-time WebSearch, each with a 2-sentence summary and source attribution. Every evening, a recap highlights what developed throughout the day and previews tomorrow's key events. Breaking news alerts fire automatically every 2 hours during daytime whenever a major story breaks — earthquakes, market crashes, political announcements — so users never miss what matters. Users can tune their experience by setting topic preferences — weighting finance over entertainment, or boosting international coverage — so every briefing reflects what actually matters to them. Supports Chinese and English output. Deliverable via Telegram, Feishu, Slack, or Discord. No registration required for on-demand queries; optional user profile unlocks personalized daily push and breaking alerts. Trigger words: 早报, 晚报, 今日新闻, 新闻摘要, 热榜, 热搜, 追踪, 最新消息, 突发, 微博热搜, 知乎热榜, X热帖, 科技新闻, 财经新闻, AI早报, AI最新, 人工智能动态, 军事新闻, 军事动态, 头条, 订阅新闻, morning briefing, daily news, news summary, Chinese news, trending, breaking news, news push, hot topics, topic tracking, international news, AI news, military news.
As a process D 47/100 · Unfinished process — weak spots: result and completion, inputs and preconditions, failures and branches
How to improve
- Shorten the description to 1024 characters.
- Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
- A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.
Guard findings · 0
✓ No critical or high findings
Files scanned: 12. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- error
description-longdescription is 1654 chars, limit 1024 - warning
name-formatname should be kebab-case (lowercase letters, digits, hyphens) - note
description-budgetdescription takes 1654 of the ~15000-char shared budget for all skills - note
frontmatter-keyunknown frontmatter key "keywords"
Process rating: all ten parameters 47/100
- 0Result and completion. Does not say what the result is
- 0Inputs and preconditions. Does not say what the process needs to start
- 0Failures and branches. Linear process with no failure handling
- 0Progress reporting. Says nothing while it works
- 40Consistency. Frontmatter name (NewsToday) differs from the folder (newstoady)
- 60Tools and files. Uses tools (web) that frontmatter does not declare
- 70When it triggers. States when to use, but not when not to
- 100Steps. 14 steps
- 100Execution cost. Instruction body is 570 tokens
- 100Running it twice. No mutating operations
Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.
Quality signals
- +5Description has no quoted example phrases that should trigger the skill
- +4Description does not say when NOT to use the skill (false activations)
- +3Description length 1653: 120–800 characters recommended
- +3Output format is not stated: the model decides each time
- +1No license
- +2Single-language instructions
- +4Structure: 14 headings
- +3Step-by-step instructions: 14 items
- +4Has examples (1 code blocks)
- +3All 7 scripts are documented
Quality base 70; lint remarks subtract, signals add up to 100. Result: 50.