AgentStack
SKILL verified MIT Self-run

Content Marketing

skill-busyeugene-content-marketing-skills-content-marketing · by busyeugene

Use when the user describes a content marketing goal at a high level (\"publish a post on X\", \"launch a campaign about Y\", \"build next quarter's calendar\") and isn't sure which specific skill to use. Routes the request to the right downstream skill or chains multiple skills into an end-to-end flow.

No reviews yet
0 installs
4 views
0.0% view→install

Install

$ agentstack add skill-busyeugene-content-marketing-skills-content-marketing

✓ scanned · ✓ verified — works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

Are you the author of Content Marketing? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

Content Marketing — Router

Optional orchestrator for the Content Marketing Skills library. Interprets a high-level goal and either dispatches to one downstream skill or chains several into a complete flow. All downstream skills work standalone; this router just saves you from remembering which to call when.

When this skill fires

Trigger this skill when the user says things like:

  • "Write me an article about {topic}"
  • "Launch a content campaign for {product feature}"
  • "Plan next quarter's content"
  • "Turn this blog post into everything" (repurpose flow)
  • "Audit our content and tell me what to fix"

Don't fire this skill when the user names a specific task ("draft editorial guidelines", "run an SEO audit") — let the specific skill handle it directly.

Setup

No keys of its own. Downstream skills check their own setup.

Routes

The router recognizes these flows. For each, it confirms the plan with the user before running, then invokes skills in order. Each step's output becomes the next step's input.

Flow A — "Write me an article about X"

Skills: content-briefblog-postproofreading-style-check

  1. If editorial-guidelines.md is missing at the project root, pause and run editorial-guidelines first.
  2. Run content-brief with the user's keyword or topic → produces briefs/.md.
  3. Run blog-post against the brief → produces posts/.md.
  4. Run proofreading-style-check against the draft → produces review/-report.md.
  5. Stop. Report the three paths and the review verdict.
  6. If the user then says "publish it", run the next flow.

Flow B — "Publish this draft"

Skills: (optional) proofreading-style-checkpublish-webflow or publish-wordpress

  1. Ask which CMS. If not obvious from env vars, use AskUserQuestion.
  2. If the draft hasn't been reviewed yet, run proofreading-style-check first and require zero critical findings before publishing.
  3. Run the chosen publish skill in draft mode by default.
  4. Report the post URL.

Flow C — "Launch a campaign about X"

Skills: content-briefblog-postproofreading-style-checklanding-page-copyemail-copysocial-postrepurpose-linkedin-carouselrepurpose-threads-x

  1. Confirm the campaign anchor: usually a new blog post or a landing page.
  2. Run the anchor skill first (blog-post or landing-page-copy).
  3. Run the downstream copy skills in parallel where possible (email, social, carousel, thread).
  4. Report every file created with a one-line "what to do with it".

Flow D — "Plan next quarter's content"

Skills: competitor-analysiscontent-calendar

  1. If a recent competitor analysis (< 30 days old) already exists under competitors/, reuse it. Otherwise run competitor-analysis fresh.
  2. Run content-calendar with the competitor data.
  3. Offer to run content-brief for each must-write entry in batch (user opts in per entry).

Flow E — "Audit our content"

Skills: seo-auditproofreading-style-check (optional) → content-calendar (for refresh queue)

  1. Run seo-audit against the domain.
  2. For each declining or critical page flagged in the audit, optionally run proofreading-style-check against its local copy if the user has one, or suggest adding it to the refresh queue.
  3. Run content-calendar in "refresh-only" mode to schedule the fixes.

Flow F — "Turn this blog post into everything"

Skills: social-post + repurpose-linkedin-carousel + repurpose-threads-x + email-copy (newsletter issue)

  1. Take one blog post path as input.
  2. Run all four skills in parallel.
  3. Report the bundle of artifacts produced.

Process (general)

  1. Classify intent. Read the user's request and match it to one of the flows above. If ambiguous, ask via AskUserQuestion with the flows as options.
  2. Preflight check. Confirm editorial-guidelines.md exists; if not, propose running the interview first.
  3. Confirm the plan. Before running anything, present a one-block plan:
  • Which skills will run, in what order
  • What files they'll read and write
  • Which env vars need to be set (list any missing ones upfront)
  1. Run sequentially. Do not parallelize skills that write to overlapping paths. Parallelize only where noted in Flow C and Flow F.
  2. Handle failures gracefully. If a downstream skill fails, stop the chain, preserve any artifacts created so far, and report the exact failure with the suggested fix.
  3. Report. Print a summary: every file created, next logical action, any skipped steps with the reason.

Fallbacks

  • Missing downstream skill: if a skill folder is missing from ~/.claude/skills/, name which one and tell the user how to install it.
  • Missing env vars: surface the full list at the start of the plan, not mid-run.
  • User interrupts mid-flow: always save partial artifacts; never roll back.

Verification

  1. The router never writes files of its own. All outputs are produced by the downstream skills.
  2. Running the router on a flow and running the same skills manually in the same order produces identical outputs.
  3. The end-of-flow report lists every file touched, in the order it was created.

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet — be the first.

Versions

  • v0.1.0 Imported from the upstream source.