YouTube Chapters & SEO Metadata
Turn a video (or its transcript) into a complete, paste-ready YouTube publishing kit — SEO title options, an optimized description with timestamped chapters, tags/keywords, and thumbnail text ideas — so an upload ranks and is navigable. Scopes the target keyword and metadata limits FIRST, analyzes the transcript for chapter boundaries, then writes the SEO-optimized metadata kit. Doing keyword research and pulling real chapter timestamps before writing is what makes the description both discoverable and accurate, instead of clickbait that doesn't match the video.
Steps
Entry step: scope. Each step names the specialist role it wants; the full working prompt is expandable.
- SEO & metadata scopeplannerentry
target keyword, limits, sections, chapter format
Show working prompt
Lock the SEO plan BEFORE writing. Step 1: record the video (title intent, transcript path, duration) and the primary target keyword/phrase plus 2-3 secondary ones the audience would search. Step 2: capture the platform limits as hard rules: title ≤ 100 chars (front-load the keyword), description first ~150 chars are the visible/snippet zone, chapters need ≥3 markers with the FIRST at `00:00`, tags within the char budget. Step 3: define the description structure (hook line with keyword -> 2-3 sentence summary -> CHAPTERS list -> links/CTA -> hashtags). Step 4: define thumbnail-text guidance (≤4 punchy words). Write an acceptance-criteria checklist (title ≤100 chars with keyword up front, description opens with the keyword in the first line, ≥3 chapters with first at 00:00 and monotonic timestamps within duration, tag list present, 3 title options, thumbnail text ideas) and decisions to `notes/scope.md` via write_task_note AND the produces path.
- Find chapters & keywordsdata-analyst
topic-shift timestamps + keyword extraction
Show working prompt
Mine the transcript for structure and terms. Step 1: read the transcript with timestamps and detect topic boundaries where the subject clearly shifts. Step 2: for each boundary record a `MM:SS`/`HH:MM:SS` timestamp and a short keyword-aware title; force the first to `00:00`. Step 3: enforce ≥3 chapters, monotonic timestamps within the duration, and reasonable spacing. Step 4: extract the recurring topics/terms the video actually covers (frequency + relevance) to inform tags and ensure the title/description match the content. Step 5: write the chapters and a candidate keyword/tag list to `notes/analysis.json` via write_task_note AND the produces path. On a loop-back, adjust only the flagged chapters/terms.
- Write the metadata kitcopywriter
title options, description+chapters, tags, thumbnail text
Show working prompt
Assemble the paste-ready kit from the analysis. Step 1: write 3 title options, each ≤100 chars with the primary keyword front-loaded; mark a recommended one. Step 2: write the description: a hook FIRST line containing the keyword (this is the search snippet), a 2-3 sentence summary, then the CHAPTERS block as plain `00:00 Title` lines (first line `00:00`) pasted from the analysis, then a links/CTA section, then 3-5 hashtags. Step 3: write a comma-separated tags line from the keyword analysis within the budget. Step 4: give 2-3 thumbnail text ideas (≤4 words each). Step 5: assemble everything into `youtube-metadata.md` under clear headers. Keep claims accurate to the video — no clickbait that the content does not deliver. write_task_note the path and chapter count. On a loop-back, fix only the flagged sections.
- Evaluatereviewer
Grade the deliverable against every acceptance criterion. All pass → finish; any fail → loop back and fix the gap.
Show working prompt
Validate the kit against platform rules. Step 1: confirm `youtube-metadata.md` has all sections (3 titles, description, chapters, tags, thumbnail text) — PASS/FAIL. Step 2: confirm each title is ≤100 chars with the keyword front-loaded, and the description's FIRST line contains the keyword. Step 3: confirm the chapters block has ≥3 markers, the first is `00:00`, timestamps are monotonic and within the video duration. Step 4: spot-check 2 chapter timestamps against the transcript — real topic shifts? Step 5: confirm tags fit the budget and titles/description honestly match the content (no clickbait). Write PASS/FAIL per criterion. Then route — this is the whole point of the loop: - **Every criterion PASSES →** call `advance_task_step({ ref, stepId: "evaluate", next: "finish" })`. - **Any criterion FAILS →** write the specific gaps to notes, then call `advance_task_step({ ref, stepId: "evaluate", next: "write" })` to loop back. The builder fixes exactly those gaps. Never route to `finish` while any criterion is unmet. The build phase's completion gate already blocked a grossly-incomplete deliverable; your job is the judgment an automated check cannot make (does it actually work, read well, look right). After ~3 unproductive loops, stop and report DONE_WITH_CONCERNS so the user can step in. - Finishdeveloper
All acceptance criteria met. Stamp a short summary and report DONE.
Show working prompt
Every acceptance criterion passed. Write a one-paragraph DONE summary to task notes via `write_task_note`: what was built, the deliverable path(s), and a one-line confirmation that each criterion is met. Then report DONE.
Triggers
Phrases that suggest this craftbook to a crew.
- youtube description and chapters
- seo title for my video
- youtube metadata
- optimize my video description
- video tags and chapters
Source
View this craftbook on GitHub · MIT license