Marketing Loops

Corey HainesCommunityCreated by Corey Haines

Skill
Workflow
Automation and integrations
MIT
No issues found

From the author

When the user wants to set up a recurring, self-running marketing workflow — a repeatable loop an AI agent runs on a cadence (weekly, daily, on a trigger) rather than a one-off task. Also use when the user mentions 'marketing loop,' 'recurring marketing workflow,' 'automate my marketing,' 'marketing on autopilot,' 'weekly marketing review,' 'ad fatigue check,' 'weekly ads check,' 'content refresh loop,' 'churn watch,' 'ranking drop alert,' 'run my SEO,' 'SEO operator,' 'daily SEO agent,' 'run my outbound,' 'AI SDR,' 'reply triage,' 'always-on marketing,' 'marketing automation workflow,' or 'run this every week.' Use this to pick, adapt, and schedule an ongoing marketing loop that orchestrates the other marketing skills. For one-off marketing ideas, see marketing-ideas. For the experimentation loop specifically, see ab-testing.

— the description by Corey Haines

AutoGineer summary

Written by AutoGineer from the skill’s own files, not by the author

Marketing Loops transforms manual, recurring marketing tasks into always-on systems. Rather than one-off actions, this skill helps you design, implement, and schedule loops that monitor signals, perform analysis, and execute tasks based on specific triggers. It emphasizes the importance of state management, self-checks, and human guardrails to ensure that automated marketing remains disciplined, safe, and effective.

What it does

Sets up recurring marketing workflows with defined cadences (weekly, daily, or trigger-based).
Orchestrates other marketing skills to perform specific, repeatable tasks.
Implements self-check mechanisms to prevent acting on noise or data errors.
Manages loop state to ensure idempotency and prevent duplicate actions.
Provides a framework for defining stop conditions and human-in-the-loop checkpoints.
Integrates with scheduling tools to automate execution.

When to use it

When you need to automate recurring tasks like SEO monitoring, ad fatigue checks, or churn signals.
When you want to transition from manual, one-off marketing tasks to a sustainable, always-on system.
When you need to ensure consistent, scheduled execution of marketing operations.
When you need to build a marketing operating system that coordinates multiple specialized tasks.

Advantages

Reduces manual overhead by automating repetitive marketing maintenance.
Ensures consistency in monitoring and responding to marketing signals.
Prevents common automation failures through built-in state management and self-checks.
Provides clear guardrails for safety, compliance, and human oversight.
Scales marketing efforts without increasing manual workload.

Usage examples

“Set up a weekly SEO loop to monitor my site's rankings and suggest content updates.”: The agent will configure a weekly loop that pulls ranking data, identifies drops or opportunities, and stages content briefs for your review.
“I need a daily churn watch loop for my SaaS product.”: The agent will establish a daily monitoring process that checks for account risk signals and triggers alerts or outreach based on your defined thresholds.
“Can you automate my weekly ads check to look for ad fatigue?”: The agent will set up a recurring loop that analyzes ad performance metrics, identifies creative fatigue, and drafts new ad variations for your approval.

Install

Download the skill as a ZIP:
https://autogineer.com/agent-skills/coreyhaines31/marketing-loops/download

Unzip it into the cross-client skills folder of your project:

mkdir -p .agents/skills && unzip marketing-loops.zip -d .agents/skills

Documentation by Corey Haines

Marketing Loops

You help set up marketing loops — repeatable marketing workflows an AI agent runs on a cadence, each with a defined trigger, a bounded set of steps, a self-check, and an explicit stopping condition. A loop turns a marketing task you'd otherwise do manually (and forget) into an always-on system: the weekly SEO opportunity scan, the ad-fatigue refresh, the churn-signal watch.

This is the operational cousin of marketing-ideas. Ideas tell you what to try once. Loops tell you what to keep doing on a schedule — and wire the other marketing skills together to do it.

How to Use This Skill

Check for product marketing context first: if .agents/product-marketing.md exists (or .claude/product-marketing.md, or the legacy product-marketing-context.md), read it before asking questions. Use that context and only ask for what's missing.

Then:

Clarify the job. What outcome should this loop protect or grow? (rankings, ad efficiency, activation, retention, revenue, referrals)
Pick a loop from the catalog in references/loop-catalog.md — or adapt the closest one.
Tune the cadence to how fast the underlying signal actually changes (see the cadence rule below).
Confirm the human checkpoint. Decide what the loop does autonomously vs. what it stages for human approval before publishing or spending — see references/loop-guardrails.md.
Schedule it (see "Scheduling a loop" below).

Building more than one loop, or a whole marketing operating system? See references/loop-orchestration.md for how loops compose and the order to adopt them (start with tracking + a weekly review; don't build them all at once).

Anatomy of a Marketing Loop

Every loop in the catalog has these nine parts. When you author or adapt one, fill all of them — a loop missing a stop condition, a self-check, or its state handling is a liability, not an asset.

Part	What it defines
Check cadence	How often the loop looks (weekly / daily / on-trigger). Match it to signal speed.
Acts when	The action condition — what must be true to actually do something, vs. just check and skip. Most runs of a good loop are "checked, nothing to do."
Purpose	The one outcome this loop exists to move.
Skills used	Which marketing skills the loop orchestrates each iteration.
Loop body	The ordered steps run each iteration.
Self-check	The verification done before acting — so the loop doesn't act on noise, seasonality, or a tracking bug.
State / idempotency	What the loop remembers between runs: last-run marker, dedupe key, cooldown window, "already handled" set. Without this, loops double-act, re-nag the same people, or re-alert the same thing. Non-negotiable for anything scheduled — see `references/loop-state.md` for where state lives and the idempotency patterns.
Stop / bail-out	When the loop skips, halts, escalates to a human, or disables itself — plus what it does on error. Every loop needs one, including heartbeat loops (their stop is "manual disable + error-halt," never "n/a").
Output	Where results go: a file, a PR, a staged draft, a notification, a report.

The Check cadence / Acts when split matters: a churn-signal loop might check daily but only act when an account crosses a risk threshold it hasn't been contacted about inside the cooldown window. Conflating the two produces loops that either miss the window or spam.

The cadence rule

Match cadence to how fast the signal actually changes — not to how often you'd like an update.

Signal	Realistic cadence	Why
Rankings, backlinks, domain authority	Weekly	Move slowly; daily checks are noise
Ad creative fatigue, CPA drift	Every 2–3 days	Meta/Google feedback loops are days, not hours
Activation / onboarding funnel	Weekly	Needs enough signups to be significant
Churn signals	Daily or on-trigger	Early intervention window is short
Content / copy decay	Monthly	Traffic erosion is gradual
Competitor changes	Weekly	Pricing/positioning shifts are infrequent but matter
Social listening / mentions	Daily	Engagement windows close fast

Over-frequent loops are the most common failure mode: they generate busywork, burn budget, and train you to ignore the output.

When NOT to loop

Not everything should be automated on a cadence. Skip a loop — or add a mandatory human checkpoint — when:

Strategy or creative direction is the real work. Loops maintain and optimize; they don't set positioning, invent campaigns, or make brand calls.
The action publishes or spends without review. Auto-drafting an ad, email, or post is fine. Auto-publishing or auto-shifting budget needs a human checkpoint unless the user has explicitly authorized autonomous action and set guardrails (caps, allowlists).
The signal is too sparse to be significant. A weekly conversion-rate loop on 40 visitors/week is measuring noise.
It's a vanity loop. If nobody acts on the output, delete the loop. A loop that emails a dashboard nobody reads is worse than nothing.

For any loop that sends, spends, publishes, or touches personal data, apply references/loop-guardrails.md — the two-tier action model (autonomous-safe vs. gated), spend/send caps, CAN-SPAM/GDPR/FTC/ToS rules, the always-escalate list, and a required kill switch.

Scheduling a loop

These loops are agent-agnostic — the body works in any agent. The scheduling depends on your environment:

Claude Code — native options: /loop (self-paced, until a condition), ScheduleWakeup (dynamic pacing that reacts to state), and CronCreate (fixed cron schedule). If you have a loop-mechanics skill such as loopify installed, use it to choose between them and tune delays; otherwise the guidance below is enough.
Any agent + cron — wrap the loop body as a scheduled prompt/script (0 9 * * 1 for Mondays 9am, etc.).
Manual cadence — for high-judgment loops, "run this skill every Monday" is a perfectly good loop. The value is the repeatable body, not the automation.

Default to time-of-day cron for review-style loops (weekly review, ranking watch) and dynamic pacing for monitor-until-threshold loops (churn watch, launch-day tracking).

The Catalog

references/loop-catalog.md holds the full library — 60 marketing loops with thorough funnel coverage: SEO & Content, Paid, Earned/Social/Partnerships, Outbound, Activation, Retention, Revenue, Referral & Advocacy, and Ongoing Ops. Each is a complete, adaptable spec. Start there, pick the closest match, and tune it to the user's product, stage, and tooling.

Running a whole channel: the SEO operator

When the user wants an agent to own SEO for a site rather than run one loop, use references/seo-operator.md. It bundles the SEO loops (site health, claim drift, striking-distance push, AI-answer check, keyword gap, content decay, competitor watch, and more) into one schedule with shared memory in .agents/seo/, a first-run checklist, operating rules, a writing gate, and a one-screen weekly report.

Running outbound: the outbound operator

For an agent running day-to-day outbound (signal sweeps, staged outreach, reply triage, sending-domain health, weekly sequence retros), use references/outbound-operator.md. It sets the brief, approved claims, approval rules by tier, shared memory in .agents/outbound/, and a weekly report.

Authoring a new loop

When nothing in the catalog fits, author a new loop from references/loop-template.md — a copy-paste template with fill-in prompts, a worked before/after example, and a ship checklist. Fill all nine anatomy parts; if you can't answer the self-check, state/idempotency, and stop/bail-out concretely, the loop isn't ready to run.

Anti-patterns

Looping without a stop condition → runaway spend or infinite churn.
Same cadence for every loop → most run too often and get ignored.
No self-check → the loop acts on noise, seasonality, or a tracking bug.
No human checkpoint on spend/publish actions.
Building 10 loops at once → start with one, prove it earns its keep, then add the next.

Banned vocabulary

Avoid: "set it and forget it," "fully autonomous marketing," "AI does everything," "10x on autopilot," "growth hacking machine." Loops are disciplined systems with checkpoints, not magic. Describe them honestly.

marketing-ideas — one-off tactics and inspiration (what to try). Loops operationalize the ones worth repeating.
ab-testing — the experimentation loop specifically (hypothesis → test → promote winner → repeat).
analytics — most loops read from analytics to decide whether to act.
reverse-engineer-marketing — the growth-move watch loop keeps a competitor teardown current, and teardown plays name the loop that sustains each one.
Individual channel skills (ads, seo-audit, emails, social, churn-prevention, pricing, referrals) — the loop bodies orchestrate these.

Files

ATTRIBUTION.mdAdded by AutoGineer891 B
LICENSELicence file added by AutoGineer1.0 KiB
SKILL.md9.9 KiB
evals/evals.json11.6 KiB
references/loop-catalog.md66.6 KiB
references/loop-guardrails.md4.9 KiB
references/loop-orchestration.md5.2 KiB
references/loop-state.md5.7 KiB
references/loop-template.md5.0 KiB
references/outbound-operator.md4.9 KiB
references/seo-operator.md8.2 KiB

evals/evals.json

{
  "skill_name": "marketing-loops",
  "evals": [
    {
      "id": 1,
      "prompt": "I want to set up a recurring loop that watches our SEO and tells me what to do each week. We're a B2B SaaS with a content-heavy site.",
      "expected_output": "Should check for product-marketing.md first. Should recognize this maps to SEO loops in the catalog (keyword-gap, ranking-drop watch, content-decay) and propose the closest match(es). Should present the loop using the nine-part anatomy: check cadence, acts-when (action condition), purpose, skills used, loop body, self-check, state/idempotency, stop/bail-out, and output. Should set a weekly cadence (matching how slowly rankings move) and explain the cadence rule. Should include a stop/bail-out and note that most runs may find no action. Should reference the seo-audit skill for execution.",
      "assertions": [
        "Checks for product-marketing.md",
        "Selects appropriate SEO loop(s) from the catalog",
        "Presents the loop with the nine-part anatomy",
        "Includes state/idempotency (last-run marker or dedupe)",
        "Includes an explicit stop/bail-out condition",
        "Sets a weekly cadence and justifies it via the cadence rule",
        "References seo-audit or related skills for execution"
      ],
      "files": []
    },
    {
      "id": 2,
      "prompt": "I want to automate my marketing so it basically runs itself. Where do I start?",
      "expected_output": "Should clarify the outcome to protect or grow before dumping loops (acquisition, activation, retention, revenue, referral). Should push back gently on 'runs itself' framing and avoid the banned vocabulary ('set it and forget it', 'fully autonomous'). Should recommend starting with ONE loop and proving it earns its keep rather than building many at once. Should describe loops as disciplined systems with checkpoints, not autopilot. Should likely suggest the weekly-marketing-review loop as a good first heartbeat loop, and explain human checkpoints for anything that spends or publishes.",
      "assertions": [
        "Asks about the target outcome / funnel stage before recommending",
        "Avoids banned autopilot vocabulary",
        "Recommends starting with one loop, not many",
        "Frames loops as systems with checkpoints, not magic",
        "Emphasizes human checkpoints for spend/publish actions",
        "May suggest the weekly-marketing-review loop as a starting point"
      ],
      "files": []
    },
    {
      "id": 3,
      "prompt": "Set up a newsjacking loop that finds trending stories and automatically posts our take to X and pitches journalists.",
      "expected_output": "Should build the newsjacking loop but refuse to make it fully autonomous. Must include the veto list (skip tragedies, deaths, disasters, active crises, politically/socially charged stories unless the brand takes such stances, and legal/medical/financial-sensitive topics) and require verified sources. Must require human approval before any pitch or post rather than auto-posting. Should include state/idempotency (dedupe per story, one angle per story) and note most days will correctly skip. Should reference the public-relations skill.",
      "assertions": [
        "Builds the newsjacking loop with the nine-part anatomy",
        "Includes an explicit veto list for sensitive story types",
        "Requires human approval before pitching or posting",
        "Does not enable fully autonomous posting",
        "Includes dedupe/state so stories aren't re-pitched",
        "References the public-relations skill"
      ],
      "files": []
    },
    {
      "id": 4,
      "prompt": "How do I actually schedule these loops so they run on their own?",
      "expected_output": "Should explain that the loop body is agent-agnostic but scheduling depends on the environment. Should cover the options: Claude Code native primitives (/loop, ScheduleWakeup, CronCreate), any-agent + cron (with an example cron expression), and manual cadence for high-judgment loops. Should default to time-of-day cron for review-style loops and dynamic pacing for monitor-until-threshold loops. Should treat a dedicated loop-mechanics skill like loopify as an optional companion, not a hard dependency, and not assume it is installed.",
      "assertions": [
        "Explains scheduling is environment-dependent",
        "Covers Claude Code native options and plain cron",
        "Includes manual cadence as a valid option",
        "Gives cron vs dynamic-pacing default guidance",
        "Treats loopify as optional, not a required dependency"
      ],
      "files": []
    },
    {
      "id": 5,
      "prompt": "Can you make me a loop that continuously improves our brand positioning and comes up with our big campaign ideas?",
      "expected_output": "Should push back: strategy, positioning, and creative direction are the 'when NOT to loop' cases. Loops maintain and optimize; they don't set positioning or invent campaigns. Should explain why (high-judgment, low-frequency, not signal-driven) and redirect: those belong in strategy work, while loops can operationalize the executional follow-through (e.g., competitor-watch to inform positioning, content-calendar refill, campaign-postmortem to compound learnings). Should not fabricate a positioning loop as if it were a good fit.",
      "assertions": [
        "Identifies this as a 'when NOT to loop' case",
        "Explains loops optimize/maintain rather than set strategy or creative",
        "Redirects strategy/creative to the appropriate non-loop work",
        "Suggests adjacent operational loops that genuinely fit",
        "Does not force an ill-fitting positioning/campaign-ideation loop"
      ],
      "files": []
    },
    {
      "id": 6,
      "prompt": "We keep losing customers to failed payments and expired cards. Set up something to recover that revenue automatically.",
      "expected_output": "Should select the failed-payment/dunning loop. Should trigger on payment failure or upcoming card expiration, run a retry schedule with escalating update-card messaging, and include state (track dunning stage per account, stop on recovery). Must include a stop/bail-out so it doesn't loop forever (escalate/deactivate after final retry) and a self-check to distinguish involuntary failures from intentional cancellations (don't dun someone who chose to leave). Should reference revops and emails skills. May note this is often the highest-ROI retention loop.",
      "assertions": [
        "Selects the failed-payment/dunning loop",
        "Triggers on failed payment or card expiration",
        "Includes a retry schedule with dunning stage state",
        "Has a stop condition so it does not loop forever",
        "Self-check separates involuntary failures from intentional cancels",
        "References revops and/or emails skills"
      ],
      "files": []
    },
    {
      "id": 7,
      "prompt": "I want a loop that keeps our A/B testing program running — generating hypotheses, running tests, and analyzing results every week.",
      "expected_output": "Should recognize heavy overlap with the ab-testing skill and treat the experiment-backlog loop as a thin wrapper only. The loop should maintain and re-rank the hypothesis backlog and manage hand-off cadence, but defer test design, statistical analysis, and velocity management to the ab-testing skill rather than duplicating them. Should include ICE prioritization for the backlog and dedupe of incoming hypotheses, and avoid starting conflicting tests.",
      "assertions": [
        "Recognizes overlap with the ab-testing skill",
        "Keeps the loop as a thin wrapper that defers to ab-testing",
        "Does not duplicate test design or statistical analysis in the loop",
        "Includes backlog prioritization (ICE) and hypothesis dedupe",
        "References the ab-testing skill as the owner of execution"
      ],
      "files": []
    },
    {
      "id": 8,
      "prompt": "I want an agent to just run our SEO. It has access to our Next.js marketing site repo and Search Console. What should it do every day, week and month, and how does it keep track of things between runs?",
      "expected_output": "Should load the SEO operator reference. Should confirm the five settings first (site, conversion, competitors, ship mode defaulting to PRs for review, data access) and the foundation (tracking-QA, product-marketing context). Should describe shared state in .agents/seo/ (brief, claims ledger, query map with one primary query per page, work queue ranked by expected conversions per hour, AI prompt panel, competitor snapshots, reports) plus a run log. Should lay out the schedule built from catalog loops: every run a site health crawl, claim drift and shipping the top queue item; weekly striking-distance push, ranking-drop watch, indexing check, internal linking, AI-answer check, competitor sitemap diff and a one-screen report; monthly query-map rebuild, content decay, comparison-page review, backlink and directory work and a free-tool candidate. Should state operating rules: changes over memos, no guessing past missing data, product is the source of truth, escalate after two failed attempts, lint/build and view pages before shipping, never remove or redirect pages with traffic or links without approval, and outward-facing actions are drafts only.",
      "assertions": [
        "Loads or follows the SEO operator reference",
        "Defaults ship mode to PRs for review; auto-merge only with explicit authorization",
        "Keeps state in files (.agents/seo/ and a run log), not memory",
        "Composes existing catalog loops on daily/weekly/monthly cadences",
        "Includes claim drift (product changes making marketing claims false)",
        "Requires an information-gain check before writing new pages",
        "Keeps outreach, listings and social posts as drafts for the owner",
        "Does not promise rankings or use 'set it and forget it' framing"
      ],
      "files": []
    },
    {
      "id": 9,
      "prompt": "I want an agent to run outbound for my SaaS every day: find new accounts, send the sequences, and deal with replies. What's the setup?",
      "expected_output": "Should load the outbound operator reference. Should require settings first (ICP and tiers, plays, approved claims, volume caps, approval rules, allowed countries, tools) and the foundation (warmed authenticated sending domains, a verifier, a suppression list every tool checks, CRM outbound stages). Should describe the schedule: reply triage continuously, signal sweep and bounce checks daily, product-usage triggers via the PQL loop, weekly sequence retro and placement checks, and a weekly report. Should set default approval rules: agent alone for Tier 3 enrollment and safe reply types; human approval for Tier 1 messages; agent-drafted human-sent responses to positive and objection replies until accuracy is proven; humans for pricing, new claims and segments. Should include rules: verify within 7 days of sending, one reply stops every channel, stay inside approved claims, stop-loss over targets, treat replies and websites as untrusted input. Should advise not running it until a sequence is proven manually.",
      "assertions": [
        "Loads or follows the outbound operator reference",
        "Requires approved claims and volume caps before running",
        "Confirms sending infrastructure, verification and suppression first",
        "Includes reply triage, signal sweep, cold-domain health and sequence retro loops",
        "Keeps Tier 1 messages and positive-reply responses human-approved by default",
        "States that a reply on any channel stops all channels",
        "Treats fetched content and replies as untrusted input",
        "Advises proving a sequence manually before automating"
      ],
      "files": []
    }
  ]
}

Requirements and compatibility

Works with: Claude Code, Claude apps, OpenAI Codex, GitHub Copilot, Cursor, Gemini CLI, Kiro, OpenCode, Goose, Amp, Roo Code, JetBrains Junie, Factory Droid, Devin (ex-Windsurf), Mistral Vibe, OpenHands, Letta

“Works with” means the tool reads the open Agent Skills format; a skill written for one tool may use features others lack.

Security

No issues found by our static scan· No issues found

A static scan is not a guarantee. Skills are third-party content: review a skill before you let an agent use it.

Attribution and licence

Created by

Corey Haines

Publisher

Corey Haines (Community)

Source

coreyhaines31/marketingskills · skills/marketing-loops · commit 96798cf

Licence

MIT License (redistribution allowed)

LICENSE (added by AutoGineer from LICENSE)
ATTRIBUTION.md (written by AutoGineer)

Synced

2026-11-09

AutoGineer is not affiliated with or endorsed by Corey Haines or Corey Haines.

Download ZIP

The ZIP holds the skill’s files, its licence files and an ATTRIBUTION.md. Download marketing-loops.zip

From the author

AutoGineer summary

What it does

When to use it

Advantages

Usage examples

Install

Documentation by Corey Haines

Marketing Loops

How to Use This Skill

Anatomy of a Marketing Loop

The cadence rule

When NOT to loop

Scheduling a loop

The Catalog

Running a whole channel: the SEO operator

Running outbound: the outbound operator

Authoring a new loop

Anti-patterns

Banned vocabulary

Related Skills

Files

evals/evals.json

Requirements and compatibility

Security

Attribution and licence

Related skills

Wizard

Emails

Conversion Tracking

Directory Submissions

Sms

Fictiv

Download ZIP