| name | post-retro |
| description | Run the 72-hour retrospective on a published draft and save the conclusion in retros/. Use when the user says "retro this post", "review how this performed", or wants to update the draft lifecycle after publishing. |
post-retro
Run this 72 hours after a draft is published.
Inputs to read first
- The published draft in
drafts/YYYY-MM-DD-<slug>.md
bun run post-patterns- Any known comment summary or performance notes the user provides
Preconditions
The draft frontmatter should include:
published_urlpublished_atimpressions_24himpressions_72hlikes_72hcomments_72hshares_72h
If these are missing, ask for the missing metrics or tell the user exactly which fields still need to be filled in.
Questions to answer
Every retro must answer:
- Did the post beat median impressions?
- Did it outperform similar
topic_family + source_type posts?
- Did comments validate the intended discussion angle?
- Was the hook accurate to the body?
- Should this pattern be repeated, modified, or blocked?
Output artifact
Save one markdown file to retros/YYYY-MM-DD-<slug>.md with YAML frontmatter:
---
draft_file: drafts/YYYY-MM-DD-<slug>.md
topic_family: security
source_type: news
published_url: ...
published_at: ...
impressions_24h: 1200
impressions_72h: 1800
likes_72h: 45
comments_72h: 7
shares_72h: 3
beat_median_impressions: true
beat_peer_group: false
discussion_validated: true
hook_matched_body: true
decision: modify
summary: Strong reach, but the comment thread drifted away from the intended angle.
---
The body should explain the decision in a few short paragraphs and name one concrete thing to repeat, modify, or stop.
