Squeeze every published winner — repost proven posts with a fresh hook, re-share your own winners with added context, and fan one outlier post into multiple new formats. Use on "what should I repost", "recycle my winners", "this post popped, what now", "turn this outlier into more posts", "wring the towel", after an engagement pull shows a post outperforming baseline, or on the Friday review's recycle step. Reads published bundles + engagement data; stages new drafts; never auto-posts.
インストール
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
Squeeze every published winner — repost proven posts with a fresh hook, re-share your own winners with added context, and fan one outlier post into multiple new formats. Use on "what should I repost", "recycle my winners", "this post popped, what now", "turn this outlier into more posts", "wring the towel", after an engagement pull shows a post outperforming baseline, or on the Friday review's recycle step. Reads published bundles + engagement data; stages new drafts; never auto-posts.
Consulting Content Recycler
Most operators find a winner, use it once, and move on. This skill is the reuse loop: a winner is
an asset, and an outlier is information. Only a fraction of followers saw the original; everyone
who followed since has never seen it; the author gets bored of their own material long before the
audience does. Origin evidence: swipe/posts/2026-07-09-anson-lin-founder-content-system.md.
This is notconsulting-content-distribution (publish-day spread of one new pillar across
channels). The recycler works the back catalog on a cadence.
Steps
Find the winners. Read content/04-published/ bundles + engagement data in
integrations/linkedin/engagement/ (pull fresh if stale:
python integrations/linkedin/_work/pull_posts.py --max 25 --yes, then
pull_engagement.py --post-url <url> for candidates). A winner performs clearly above the
account's recent baseline; an outlier is far above it. No engagement data → pull it, don't guess.
Diagnose why it worked. An outlier is information — isolate the variable before reusing it:
the hook pattern (grade via consulting-linkedin-hooks), the format (which founder format /
archetype — consulting-linkedin-post-architect references), the topic, or who showed up in the
comments (ICP fit? → that's also a consulting-linkedin-audience lead moment). Write the
diagnosis down in the bundle's distribution.md — it steers the whole pipeline, not just this post.
Pick the play (or several):
Repost — after 1–2 months (or a follower-growth spurt), run the same idea again: rewrite the
hook (different pattern, same substance), refresh any numbers to current verified values, keep
the core. Stage as a new single draft. Don't repost verbatim inside ~60 days.
Re-share with context — while a post is running hot, or months later as backstory for new
followers: quote/re-share your own post with an update or added context. On X, quote-post views
stack onto the original's distribution (per the source playbook); on LinkedIn, a
reshare-with-thoughts resurfaces it to the newer audience. The ceiling of this move is the
milestone chapter: quoting your own old post as un-fakeable proof of the arc.
Fan out the outlier — one winner is never one post. Re-run the same material through other
founder formats: a popped failure story returns later as a contrast hook, a milestone chapter
once things turn, a build log for what came next. A launch is at least three posts sitting in
one event: the announcement, the updated numbers, the lessons.
Stage, never send. Each recycled piece becomes a normal draft in content/03-drafts/
(single or bundle per content/AGENTS.md), body via consulting-copy-writer, structure via
consulting-linkedin-post-architect, shipped only through consulting-linkedin-publisher.
Record lineage. In the new draft's frontmatter/meta note recycled_from: (the original bundle
path + post URL) and log the recycle in the original bundle's distribution.md. Lineage prevents
accidental double-recycling and lets the Friday review see which winners are already worked.
Rails
Verified numbers only. "Update the numbers" means re-verify against the primary source
(evidence discipline) — never inflate or estimate a receipt.
Never auto-post. Everything stages as a draft; publishing stays a human action.
Voice floor:consulting-copy-writer on every rewrite — a recycled post is a new post.
Don't strip-mine. One recycle play per winner per cycle; the feed should never read as reruns.
Quality bar unchanged: a cycle with nothing worth recycling produces nothing.
Cadence
Fold into consulting-friday-review (scan for winners/outliers) or run on demand when a post pops.
Chains: consulting-linkedin-audience (the same hot post is also a lead source) →
consulting-followup-sequencer for re-engaged contacts.