| name | track-awesome-yesterday-analyzer |
| description | analyze yesterday's updates on trackawesomelist by starting from https://www.trackawesomelist.com/, retrieving the actual updated awesome lists and their newly added items, then producing a long-form structured chinese report. use when the user asks to analyze yesterday's track awesome list, yesterday's awesome list updates, recent trackawesomelist changes, or wants a deep report that first covers all updates and then selects the most important ones for deeper interpretation. |
Track Awesome Yesterday Analyzer
Use this skill to turn Track Awesome List's yesterday update stream into a structured Chinese analysis report. Start from the homepage URL the user provides or from https://www.trackawesomelist.com/. Always inspect the real updated content from yesterday rather than guessing from memory or giving a curated summary only.
Workflow
- Open Track Awesome List and locate the section that corresponds to yesterday's updates.
- Enumerate every awesome list updated yesterday.
- Open each updated list page and extract the newly added entries for that day.
- Build a complete update inventory first.
- Select the few additions that are genuinely most worth deeper analysis.
- Write a long-form Chinese report using the template in
references/report_template.md.
Retrieval rules
- Always use live web retrieval. Do not rely on stale memory for yesterday's content.
- The user input is just the homepage URL. You must navigate from there and discover yesterday's updated lists yourself.
- Yesterday means the calendar day immediately before the current date in the user's timezone.
- The first section of the report must cover yesterday's actual updates broadly, not a hand-picked subset.
- Distinguish between:
- the awesome list that was updated
- the newly added items inside that list
- Prefer the newly added items and the surrounding list context as the factual basis of analysis.
- If a list page is ambiguous, verify with the corresponding GitHub diff, commit view, or rendered list page before summarizing.
Selection rules for deep analysis
After completing the full inventory, pick a limited set of additions for deep analysis. Use these signals:
- appears across an important tool or workflow shift
- reveals a new category forming or accelerating
- is unusually actionable for builders, product engineers, or technical decision makers
- has strong implications for ecosystem direction, not just novelty
Do not simply pick the most famous project names. Choose the additions with the strongest explanatory value.
Writing rules
- Write in Chinese.
- Produce a long-form, structured report, not short bullets.
- Keep the opening overview concise, then expand the analysis in later sections.
- Separate factual inventory from interpretation. Make it clear what was updated and what your inference is.
- Do not call the first section “精选”. The first section must be a complete overview of yesterday's updates.
- When counts are available, include them, but do not invent exact numbers if you could not verify them.
- Use citations for factual claims whenever web sources were used.
Required report shape
Follow references/report_template.md closely.
The report must include these parts in order:
- Yesterday update snapshot
- Full update inventory by awesome list and newly added items
- Selected additions worth deep analysis
- Signal interpretation and ecosystem reading
- Practical follow-up suggestions
You may add small supporting subsections, but do not remove these five parts.
Quality bar
Before finalizing, check all of the following:
- Did the report start from yesterday's full update set rather than a curated subset?
- Did each inventory entry describe the updated awesome list and the newly added items?
- Did the deep-analysis section clearly explain why those items matter?
- Did the ecosystem section synthesize cross-list signals instead of repeating item descriptions?
- Is the output substantial enough to be read as a standalone interpretation report?
