| name | assemble-chapter |
| description | "WHAT: assembles a framework CHAPTER from the two already-rendered blog posts — combines Blog 1 (narrative) and Blog 2 (deep dive) into a chapter directory, wires the cross-links (Blog1<->Blog2, and both -> the plugin on the monorepo + the framework's skills) and litters in CTAs, producing the chapter as the publishable unit (Blog1 + Blog2 = one chapter). It does NOT rewrite the blogs' prose — only wires links + CTAs + a chapter manifest. WHEN: when you have Blog 1 + Blog 2 for a framework and need to combine them into a chapter, when wiring the cross-links/CTAs of a framework package, or when the user mentions assembling a chapter / skill2framework (any of)." |
PROMPT
You assemble a framework CHAPTER from two already-rendered blog posts. A chapter = Blog 1 (the narrative opener) + Blog 2 (the mechanical deep dive), WIRED together with cross-links and CTAs. You do NOT rewrite the blogs' prose (voice/structure is out of scope) — you only inject the link sections, the CTAs, and a chapter manifest.
Specifics (provided at dispatch)
- Framework name:
{framework_name}
- Blog 1 source (narrative):
{blog1_path}
- Blog 2 source (deep dive):
{blog2_path}
- Plugin URL (the operational face on the monorepo):
{plugin_url}
- Skill links (the rung face — the skills this framework decomposes into; name -> path/url each):
{skill_links}
- Output chapter dir:
{chapter_dir}
STEP 1 — COPY the two posts into {chapter_dir} as blog1.md and blog2.md (verbatim — do not touch their prose).
STEP 2 — WIRE THE CROSS-LINKS + CTAs (append/replace ONLY the links section + add CTA lines; never edit narrative body)
- In
blog1.md: ensure a links block exists near the top AND at the end with:
- "Deep dive — how it works: ./blog2.md"
- "Get the plugin: {plugin_url}"
- "The skills: {skill_links}" (list each)
- A clear CTA line (e.g. "Install the plugin and run it yourself: {plugin_url}").
- In
blog2.md: ensure it links back ("The story: ./blog1.md"), links the plugin ("{plugin_url}"), and ends with a CTA.
- "Litter CTAs" = a short call-to-action at each major section break is fine, but keep them short and do NOT alter the surrounding prose. CTAs point at the plugin (install/use) and the other blog.
STEP 3 — WRITE THE CHAPTER MANIFEST {chapter_dir}/chapter.md
A short index: chapter title = {framework_name}, a 2-line abstract, then the ordered contents (1. Blog 1 narrative -> ./blog1.md, 2. Blog 2 deep dive -> ./blog2.md), the plugin link, and the skill links. This manifest is what the framework skill and the nomicon will point at.
STEP 4 — REPORT (final message)
Return: the chapter dir tree, the FINAL blog1.md + blog2.md links/CTA sections (show what you injected), and the full chapter.md. Confirm: prose bodies unchanged (diff only in link/CTA lines), cross-links resolve (blog1<->blog2 relative paths exist), plugin + skill links present in all three files.
Hard constraints
- Do NOT rewrite blog prose. Only inject link sections, CTA lines, and the manifest.
- Cross-links between blog1/blog2 are RELATIVE (
./blog1.md / ./blog2.md) so the chapter dir is self-contained.
- Touch ONLY files under
{chapter_dir}. Read nothing under /home/GOD/core.