create-content
Skill1 starsChanged 3 months agoMITView on GitHub
Plan and produce a content batch — platform-native social posts, SEO articles, email newsletter issues — on the give-value-then-ask rhythm, batched for one publishing approval. Run weekly or per the strategy's cadence.
In this repository22 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- build-funnelSkill
- close-dealSkill
- connect-toolsSkill
- define-positioningSkill
- design-offerSkill
- experimentSkill
- grow-referralsSkill
- grow-seoSkill
- health-checkSkill
- launch-campaignSkill
- launch-productSkill
- manage-pipelineSkill
- measureSkill
- plan-strategySkill
- research-marketSkill
- review-monthlySkill
- run-adsSkill
- run-outboundSkill
- run-partnershipsSkill
- setup-teamSkill
- tune-revopsSkill
The file
Files
/create-content — Earn Attention Before Asking
Inputs: knowledge/strategy.md (channels + cadence), knowledge/positioning.md, knowledge/audience-profiles.md, knowledge/content-calendar.md, performance notes in knowledge/content/.
Agents: content-marketer leads; copywriter drafts; tools-operator schedules/publishes.
Method: knowledge/methods/content-and-social.md, knowledge/methods/digital-channel-strategy.md (SEO).
Steps
- Review what's working. Last batch's numbers: formats and topics that earned engagement get more; underperformers get one more fair try or the axe.
- Plan the batch. Fill
templates/content-calendar.mdfor the period: topics from customer questions/language, formats native to each platform, give/ask ratio enforced (several value pieces per ask; asks contain exactly one plain CTA). - Draft natively. Each piece crafted for its platform's format, tone, and consumption speed — no verbatim cross-posting. Two-second test on every piece: hook/image/first line carries the value.
- SEO pieces: target one real question per article, answer it better than what currently ranks, structure for scanning. Marketplace businesses: treat listing titles/tags as the SEO surface.
- Quality pass. Voice check (
standards/voice.md), claims check including the Claims Review Checklist (standards/compliance.md), positioning alignment. Where a piece makes or implies a results claim, pull real proof fromknowledge/research/social-proof.mdrather than asserting. - Batch approval. Present the whole calendar to the operator at once — the publishing gate covers the batch; approved pieces post on schedule without further asks.
- Schedule & publish. tools-operator loads the batch into the scheduler (or posts on cadence). Confirm each goes live.
- Log performance as results land; feed next batch's step 1.
Before you start
- Required inputs:
knowledge/strategy.mdapproved (content must serve a chosen channel),knowledge/positioning.mdatseeded+,standards/voice.mdBrand voice section (or house defaults if not yet defined). - Scan
knowledge/decision-log.mdfor tagscontent,channel,voice. For SEO articles, checkknowledge/research/for any/grow-seobriefs — those take priority over invented topics. - Gate check: publishing gate covers the batch (step 6); nothing posts without the batch approval ref.
If things go wrong
- Scheduler unavailable/unauthorized → publish manually on cadence via guided steps, or park the approved batch in
knowledge/pending.mdwith its schedule. - Batch approval doesn't come before the first slot → nothing posts; slide the calendar rather than posting unapproved.
- A piece's platform rejects it (policy, format) → adapt per platform rules or swap in the next calendar item; log the rejection reason in
knowledge/content/.
Outputs
An approved, scheduled content batch; updated rolling calendar. Content batches live in knowledge/content/YYYY-MM-<slug>/. Close with the report per templates/skill-report.md.
Writeback
knowledge/content-calendar.md; drafts + results to knowledge/content/; approval to knowledge/approvals-log.md; content decisions (tag: content) to knowledge/decision-log.md.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

