Blog Embargo
/install blog-embargo
Blog Embargo Skill
Why This Exists
The blog uses a publishAt frontmatter field to support soft embargoes: post deploys to Vercel immediately on merge but is hidden from the blog index (/blog listing) until publishAt passes. The index revalidates hourly.
Without this field, posts are indexed immediately on deploy. If social posts aren't already in Buffer before the merge, you get a window where the post is live but no social amplification is queued.
This skill coordinates the three moving parts:
publishAtin MDX frontmatter (blog index gate)--publish-atinconvert-to-mdx.py(how to set it)- Buffer scheduling at
publishAt + 15 min(social timing)
The Rule
Blog index visibility and social posts must be coordinated. Always decide
publishAtbefore raising the PR.
| Scenario | publishAt | Buffer timing |
|---|---|---|
| Want coordinated launch (blog + social together) | Set to future slot | Schedule Buffer at publishAt + 15 min |
| Post can go live now, social later same day | Set to now + 30 min | Schedule Buffer at publishAt + 15 min |
| No embargo needed (emergency fix, no social) | Omit or set to now | Post immediately |
Step-by-Step
1. Decide the publish time
Before raising the PR, decide a specific AEST datetime:
- Check Buffer queue for conflicts (use Buffer dashboard or GraphQL query)
- Align with content calendar (LinkedIn 10am AEST, X 8-9am AEST preferred slots)
- Minimum: now + 30 min (enough time to schedule Buffer before deploy)
publishAt = "2026-04-04T09:00:00+11:00" # AEDT
publishAt = "2026-04-04T09:00:00+10:00" # AEST
2. Set publishAt in MDX
Pass --publish-at to convert-to-mdx.py:
python3 scripts/blog/convert-to-mdx.py \
--input projects/blog-pipeline/converted/\x3Cslug>.md \
--output projects/blog-pipeline/converted/\x3Cslug>.mdx \
--slug \x3Cslug> \
--title "\x3Ctitle>" \
--description "\x3Cdescription>" \
--tags "\x3Ctags>" \
--author "Nissan Dookeran" \
--publish-at "2026-04-04T09:00:00+11:00"
Check: MDX frontmatter should contain publishAt: "2026-04-04T09:00:00+11:00".
3. Raise PR and merge
Post deploys immediately. Direct URL (/blog/slug) resolves. Blog index (/blog) hides the post until publishAt.
4. Schedule Buffer at publishAt + 15 min
After merge, schedule social posts with dueAt = publishAt + 15 min:
# X/Twitter — at publishAt + 15 min
node scripts/buffer-post.mjs \
--text "$(cat projects/social-growth/thread-\x3Cslug>.md | ...)" \
--channel twitter \
--publish-at "2026-04-04T09:15:00+11:00"
# LinkedIn — same time or later slot
node scripts/buffer-post.mjs \
--text "$(cat projects/social-growth/linkedin-\x3Cslug>.md | ...)" \
--channel linkedin \
--publish-at "2026-04-04T09:15:00+11:00"
Or use the batch file approach with scripts/buffer-post.mjs --file posts.json.
5. Verify
- Direct URL resolves immediately after merge ✅
- Blog index shows post only after publishAt ✅
- Buffer shows posts scheduled at publishAt + 15 min ✅
Recovery: Post Already Live Without publishAt
If a post was merged without publishAt and is already indexed:
- The embargo window is gone — post is already visible
- Schedule social posts for the next appropriate slot (same day if possible)
- Don't backdate social posts to before the post went live
- Note the gap in the daily memory log
This happened with:
ollama-embeddings(2026-04-03) — live, no social queuedportkey-patterns(2026-04-03) — live, no social queued
How publishAt Works (Technical)
publishAtis an ISO8601 field added to MDX frontmatter- The blog index page (
/blog) filters posts wherepublishAt > now - The
export const revalidate = 3600means the index updates hourly - Direct slug URL (
/blog/slug) always resolves — no gate at the page level convert-to-mdx.pysets bothpublishAtand aligns thedatefield when--publish-atis provided
Integration Points
- Playbook:
playbooks/blog-publish/PLAYBOOK.md— Step 0b (Embargo Gate) uses this skill - Script:
scripts/blog/convert-to-mdx.py—--publish-atflag - Script:
scripts/buffer-post.mjs—--publish-atflag orpublishAtin batch JSON - Buffer channels: X
69c29939af47dacb694d3d1f, LinkedIn69c29382af47dacb694d24b4
Lessons Learned
2026-04-03: Two posts (ollama-embeddings, portkey-patterns) merged without publishAt. Both immediately indexed. Social copy existed but hadn't been approved yet, so nothing was in Buffer. Result: posts live with no social amplification window. Fix: this skill + mandatory Step 0b in blog-publish playbook.
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install blog-embargo - After installation, invoke the skill by name or use
/blog-embargo - Provide required inputs per the skill's parameter spec and get structured output
What is Blog Embargo?
Coordinate publishAt embargo timing across blog index visibility and social post scheduling. Ensures blog posts don't appear in the index before social posts... It is an AI Agent Skill for Claude Code / OpenClaw, with 59 downloads so far.
How do I install Blog Embargo?
Run "/install blog-embargo" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Blog Embargo free?
Yes, Blog Embargo is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Blog Embargo support?
Blog Embargo is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Blog Embargo?
It is built and maintained by Nissan Dookeran (@nissan); the current version is v1.0.1.