pipeline-content
Skill0 starsChanged 2 months agoMITView on GitHub
The social content pipeline — mkt-strategist (calendar/angles) to mkt-content (format skill) to the mkt-editor gate to the human publish checklist. Use to run or explain the end-to-end chain for hooks, reels, LinkedIn posts, or carousels.
In this repository20 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- global/CLAUDE.mdCLAUDE.md
- copywriting-frameworksSkill
- design-mdSkill
- format-carouselSkill
- format-hooksSkill
- format-linkedinSkill
- format-reelsSkill
- funnel-designSkill
- humanizationSkill
- launch-playbooksSkill
- offer-designSkill
- pipeline-brandSkill
- pipeline-campaignSkill
- pipeline-copySkill
- pipeline-landing-pageSkill
- pipeline-luxoSkill
- pipeline-refino-uiSkill
- pipeline-sistemaSkill
- scroll-worldSkill
The file
Files
Pipeline: Content
Chain for social content (hooks, reels, LinkedIn posts/carousels) from calendar direction to publish-ready.
Chain
mkt-strategist (calendar/angles)
→ mkt-content (format skill: format-hooks / format-reels / format-linkedin / format-carousel)
→ mkt-editor (humanize + quality gate)
→ PASS → publish checklist → HUMAN publishes
→ BLOCK → mkt-content (or mkt-strategist if the angle itself is broken) → re-enter mkt-editor
→ EXIT (2 loops exhausted) → escalate to director/userSteps
- mkt-strategist sets calendar direction: which pillar/theme, which angles, which formats, for which period — written to
brain/ops/handoffs/brief-<slug>.md. This is direction, not a finished brief per post;mkt-contenthas latitude within the stated angle. - mkt-content loads the format skill matching the requested output (
format-hooksfor a hook menu,format-reelsfor scripts,format-linkedinfor posts/newsletters,format-carouselfor carousels) and produces the deliverable per that skill's output structure. Carousel visual briefs additionally route todsg-uiin parallel. - mkt-editor loads
humanizationand the project pack'svoice.md, runs the humanization pass, then issues a verdict tobrain/ops/reports/copy-<slug>.md. - On PASS, the piece proceeds to the publish checklist below. Publication itself is always a human action — no agent in this system posts to a live channel.
- On BLOCK, the rollback agent is usually
mkt-content; if the verdict traces the failure back to the angle itself rather than the execution, escalate tomkt-strategistinstead of loopingmkt-contentindefinitely. - On EXIT, or after 2 loops, escalate to the user.
Publish Checklist (human step, after PASS)
- Format spec matches the target platform's current specs (dimensions, length, caption limits)
- Links/UTMs correct and tested
- Visual assets (if any) received from
dsg-uiand match the brief - Scheduled at the time window
mkt-content/the format skill recommended - Owner assigned for first-hour engagement (per
format-linkedin/format-hooksguidance where relevant)
When to Use
- Reach/engagement-first social content: hooks, reels/shorts scripts, LinkedIn posts and carousels, newsletters.
When Not to Use
- Direct-response assets whose sole job is conversion (sales pages, ad copy, email sequences) — use
pipeline-copy. - Paid campaign structure and budget — use
pipeline-campaign.
Done When
- The
mkt-editorverdict is PASS, logged atbrain/ops/reports/copy-<slug>.md. - The publish checklist is complete.
- A human has actually posted — this pipeline's job ends at "ready," not at "live."
Parallelization
A week's worth of independent posts (a reel, a LinkedIn post, a carousel, each on a different angle from the calendar) can run through mkt-content and mkt-editor in parallel. A derivative series (several short clips cut from one long-form piece) should stay sequential behind the source piece being finalized first — derivatives built against an unfinished source get rebuilt twice.
Common Deviations
- The angle itself is weak, not the execution — a hook menu from
format-hookskeeps failing the specificity bar across every variation. That's a signal to escalate tomkt-strategistto revisit the calendar angle, not to keep regenerating hooks against the same brief. - A carousel's copy passes but the visual brief handoff to
dsg-uistalls — the copy side does not wait on the design side to be marked done; they gate independently, and the publish checklist is what finally requires both. - Format spec drift (a platform's length/dimension limits changed since the format skill was last verified) —
mkt-contentflags this rather than silently shipping to an outdated spec; update the relevantformat-*skill once confirmed.
Notes for the Director
When this pipeline is one task among several in a plan, the task contract's Gate field should read mkt-editor, and Success criteria should include both the verdict file and the completed publish checklist — "content drafted" is not the done-state, "ready for a human to post" is.
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.

