원본 저장소의 제목, 예시, 코드, 표, 링크, 이미지를 유지해 표시합니다.
Overview
This skill is the head of the daily content chain, and it runs two modules in order. Module performance-review looks backward: it turns the numbers of yesterday's posts into lessons in the preferences. Module trend-discovery looks forward: it scores today's news and shapes the best trends into briefs for skill fountain-clip-finder. The skill itself updates the narratives first, because both modules read them.
Input
show- the show to run the loop for.- The Narratives and Editorial sections of the preferences.
Output
- One brief per advancing trend, handed to skill fountain-clip-finder.
A brief is a completed trend of module trend-discovery, carrying its share of the day's clip budget as clip_count.
- The report of the posts that wait, one time when the day's clips exist.
- Updated preferences: the narratives, and the lessons of module
performance-review.
Housekeeping
You MUST read HOUSEKEEPING.md if you haven't already.
Requirements
- An HTTP client, e.g. curl, for the Google News RSS route of module trend-discovery.
- Optional: a web search tool, and a social trend search tool such as one for X, when the machine has them.
- Fountain API.
- Skill fountain-clip-finder.
- Skill fountain-reports.
- Skill fountain-onboarding.
Process
- Update the Narratives section, because both modules read it.
- Run module performance-review to turn yesterday's posts and their numbers into lessons.
- Run module trend-discovery to score today's trends and shape the strongest into briefs.
- Hand each brief to skill fountain-clip-finder, and do not read its result - the chain
continues without this skill.
- Read the auto-render setting from the Automation section of the preferences, where on is the default.
List the day's drafts with the Social API either way, because step 4 hands the briefs on and never reads what came back. With auto-render on, render them with skill fountain-clip-producer as a clean final, so the user reviews the clip and not a description of it. When the Brand section holds no confirmed kit, render the single strongest clip first, and present it as the style proof of that skill. Render the rest only after the user confirms or corrects the proof, because a batch in the wrong look is a batch rendered twice. Leave the rendering to a render machine instead when one works this show, and it sends the report. Send the day's clips as the review-posts-simple report of skill fountain-reports either way, and let its approve note say whether approving renders a clip or sends it. Give each source label and its publish date when known. Use the episode for a source with an episode id, and use the external video title from context otherwise. Present the day's clips in the chat on the clip card of skill fountain-clip-finder, whether or not this run made them, because a day at budget still has clips the user has not seen.
Additional notes
Where a step repeats one API call over many items - the numbers of each post, the news of each subject, the drafts of each clip - the items do not depend on each other, so ask for them all at the same time.
This skill owns the two morning looks and nothing downstream. Finding moments and writing the copy is the job of skill fountain-clip-finder. Rendering is the job of skill fountain-clip-producer, run here or on a render machine that picks the drafts up from the Social API.
Auto-render off means something else has to render, so say which: the user's word in the chat, or a render machine that works this show. Run skill fountain-onboarding to schedule a render of the approved drafts when they want the day to finish without them, because a draft that nothing renders is a clip that never exists.
An empty Editorial section is not a wall: proceed, and say so plainly.
Use this skill for the recurring daily cycle. For a one-off "find me a clip about X", use skill fountain-clip-finder directly.
