agentleFS
moronireismoronireis/agent-system-partners/.claude/skills/pipeline-campaign/SKILL.md

pipeline-campaign

Skill0 starsChanged 2 months agoMITView on GitHub

The paid campaign pipeline — mkt-strategist (brief+budget) to mkt-copywriter and dsg-ui (creatives) to mkt-ads (PAUSED) to human approval to the mkt-ads monitoring loop and feedback to strategist. Use to run or explain the end-to-end paid-traffic chain.

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    No step bypassed the mkt-editor gate or the human-approval gate — both are hard stops, not suggestions.

    line 46

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Pipeline: Campaign

Chain for paid campaigns from strategic brief through creative production to a monitored, human-approved launch.

Chain

mkt-strategist (brief + budget)
  → mkt-copywriter (ad copy) + dsg-ui (creative visual specs) [parallel]
    → mkt-editor (gate on copy)
      → mkt-ads (builds campaign structure — PAUSED, always)
        → HUMAN APPROVAL (required — no agent activates spend)
          → mkt-ads (monitoring loop: health checks, performance diagnosis)
            → feedback to mkt-strategist (scale / kill / iterate angle)

Steps

  1. mkt-strategist writes a campaign brief to brain/ops/handoffs/brief-<slug>.md: objective, ICP/segment, offer reference, angles (from copywriting-frameworks), budget, and platform(s). Budget approval thresholds are defined per project pack; absent one, every activation and every budget change requires explicit human sign-off.
  2. mkt-copywriter produces ad primary text/headlines/descriptions per platform constraints; dsg-ui produces the visual creative specs in parallel. Both reference the same brief so angle and visual stay aligned.
  3. Ad copy passes through mkt-editor (the same gate as pipeline-copy) before it reaches mkt-ads. Visual creative follows the design squad's own review path.
  4. mkt-ads builds the campaign/ad-set/ad structure via the configured MCP (or documents the manual build steps if no MCP is configured). Every campaign is created PAUSED — non-negotiable, see .claude/rules/ads-safety.md.
  5. Human approval is required before any activation. mkt-ads presents the full structure (budget, targeting, creative, naming) for review; it never self-activates under any condition.
  6. Once live, mkt-ads runs the monitoring loop: scheduled health checks, performance diagnosis, anomaly alerts.
  7. mkt-ads reports findings back to mkt-strategist, who decides to scale, kill, or request a new creative/angle iteration — looping back to step 2 when a new angle is needed.

When to Use

  • Any paid acquisition campaign (Meta, Google, or another configured platform) from concept through live monitoring.

When Not to Use

  • Organic content — use pipeline-content.
  • Campaign structure changes with no new creative or strategic input — that's routine mkt-ads operation, not a full pipeline run.

Done When (per cycle)

  • Campaign structure exists, PAUSED, matching the brief.
  • Human approval is explicitly recorded before activation.
  • Monitoring cadence is running and reporting back to mkt-strategist.
  • No step bypassed the mkt-editor gate or the human-approval gate — both are hard stops, not suggestions.

Common Deviations

  • No budget threshold defined in the project pack — mkt-ads defaults to requiring human approval for any amount, per .claude/rules/ads-safety.md; it does not infer a threshold from past campaigns.
  • Creative from dsg-ui isn't ready when copy clears the editor gate — mkt-ads waits for both; it does not build the campaign structure against placeholder or draft creative.
  • A live campaign underperforms and the fix is a new angle, not a budget change — mkt-ads reports the diagnosis to mkt-strategist and the cycle re-enters at step 2 (new copy + creative), rather than mkt-ads iterating on targeting alone to compensate for a weak angle.
  • User asks for activation in the same message as campaign creation — these remain two separate, explicit steps regardless of phrasing; see mkt-ads's own boundaries.

Monitoring Cadence (guidance for step 6)

Health checks after activation should cover: delivery status (learning phase progress), spend pacing versus budget, early creative-fatigue signals, and any anomaly (spend spike, disapproval, zero delivery). Cadence and alert thresholds are set per project pack; absent a defined cadence, check at least once within the first 24 hours of activation and again before any scale decision.

Notes for the Director

When this pipeline is one task among several in a plan, split it into at least two task contracts: one ending at "structure built, PAUSED, awaiting approval" (gate: human), and a separate one for the monitoring loop once activation actually happens — they usually don't complete in the same session.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.