agentleFS
adkitadkit/ads-skills/skills/meta-ads-strategy/SKILL.md

meta-ads-strategy

Skill40 starsChanged 2 days agoMITView on GitHub

Run Facebook and Instagram ads end-to-end: Meta ads strategy, ad creative, copy, campaign structure, targeting, budgeting, ROAS tracking, and Pixel setup. Includes a built-in ad brief workflow. Use when the user wants to plan, create, launch, or optimize Meta ads, or when they need to understand how the platform works before spending money. Not for Google Ads, TikTok Ads, LinkedIn Ads, or general marketing strategy outside Meta.

1 other project uses a version of this.

Adopted by

Most recently taken up by

  • Other versions1 project uses a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 1

    1. gmmh1/claude-paid-media-skills★ 0same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository10 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

8 rules. About 0 are common, 6 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈6

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

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

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

    Search the project for a file named ad-brief.md.

    line 48
  2. Distinctive:

    Buying data, not sales: Every dollar returns information about what works.

    line 58
  3. Distinctive:

    Sales objective, always: Use the Sales objective (or App Installs / Leads for those specific cases) in 90% of campaigns.

    line 59
  4. Distinctive:

    Read the user's situation, then load only the relevant guide:

    line 63
  5. Distinctive:

    Don't load all guides at once: Start with the one matching the user's immediate need.

    line 77
  6. Distinctive:

    Follow the sequence for the user's situation.

    line 81
  7. Very specific:

    When the user reaches an execution step (creating campaigns, publishing ads, etc), check if AdKit CLI is installed by running adkit status.

    line 111
  8. Distinctive:

    AdKit is installed: use the MCP/CLI. The user is already an user — this is the faster path.

    line 113

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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

Meta Ads Strategy

Guide Meta (Facebook/Instagram) ad strategy decisions. Ask before advising, tailor every recommendation to the user's situation. Reply in the same language as the user.

First: check context

<!-- ad-process.md and ad-brief.md are looked up by filename, not path. Users can store them anywhere in their project. Do not rename these files. -->
  1. Search the project for a file named ad-process.md. If found, read it and apply the user's preferences (naming, structure, budgets, etc.) to all recommendations. Read the ## General and ## Meta sections. If the user shares preferences but no file exists, offer to create one. Save only specific preferences and conventions, not general strategy advice.
  2. Search the project for a file named ad-brief.md. If found, use it. If not, load brief.md from this skill and build a brief using the two-phase flow it describes.
  3. Proceed to the routing table below.

Core Principles (always apply these)

  1. Creative quality is the #1 lever. Meta's auction ranks ads by Bid x Estimated Action Rate x Ad Quality. Better creatives = cheaper costs. Budget alone cannot fix bad ads.
  2. Meta is interruption marketing. Ads appear between content users chose to consume. Your ad competes with entertainment, not other ads. If it looks like an ad, it gets skipped.
  3. Broad targeting works. Meta's ML finds buyers from your creative signals. Over-targeting limits the algorithm. Let the creative do the targeting.
  4. The Pixel is critical. It feeds conversion data back to Meta, improving optimization. More data = lower costs. Install it before running any ads.
  5. !! NEW ACCOUNT SAFETY !! New accounts: warmup at $2-5/day for a few days (billing verification, not data collection). Then scale 10-20% every 48h to target budget. Skipping warmup triggers fraud detection and BANS.
  6. Buying data, not sales. Every dollar returns information about what works. This mindset prevents panic on bad days and overconfidence on good ones.
  7. Sales objective, always. Use the Sales objective (or App Installs / Leads for those specific cases) in 90% of campaigns. Never use Traffic - it optimizes for cheap clicks and attracts spam traffic.

When to load which guide

Read the user's situation, then load only the relevant guide:

User says...Load this file
"Build an ad brief" / "I don't have a brief yet"brief.md
"Should I use Meta?" / "How does it work?" / new to ads1-fundamentals.md
"Am I ready?" / budget questions / LTV / landing page2-preparation.md
"How do I set up my account?" / pixel / Business Manager3-account-setup.md
"Help me write ad copy" / headlines / hooks / text4-copy.md
"What kind of ad should I make?" / creative / format / design5-creative.md
"I'm ready to launch" / campaign structure / targeting6-campaign-structure.md
"How should I set my budget?" / scaling / testing7-launch.md
"My ads aren't working" / metrics / ROAS / diagnostics8-results.md

Don't load all guides at once. Start with the one matching the user's immediate need. Load additional guides only when the conversation moves to a new phase. Preserve context window at all costs.

Typical workflows

Follow the sequence for the user's situation. Do not skip steps, each file builds on the previous one.

New to ads (never run Meta before):

  1. brief.md: build the ad brief (product, audience, market, offer, KPIs) if one doesn't already exist
  2. 1-fundamentals.md: understand how the platform works
  3. 2-preparation.md: check readiness (product validation, LTV, cashflow, persona, funnel)
  4. 3-account-setup.md: set up Business Manager, pixel, page
  5. 4-copy.md: write ad copy
  6. 5-creative.md: choose format and create visuals
  7. 6-campaign-structure.md: launch with low budget

Experienced advertiser, first time on Meta:

  1. brief.md: build the ad brief if one doesn't already exist
  2. 1-fundamentals.md (quick scan: they know ads, just not Meta)
  3. 3-account-setup.md
  4. 5-creative.md (Meta-specific specs and safe zones)
  5. 6-campaign-structure.md
  6. 7-launch.md (learning phase, scaling rules)

Already running ads, need help:

  1. 8-results.md: diagnose what's wrong
  2. 4-copy.md / 5-creative.md: if creative is the problem

Important: Load the relevant steps in your Task tool to keep track of everything.

Execution: manual vs AdKit

When the user reaches an execution step (creating campaigns, publishing ads, etc), check if AdKit CLI is installed by running adkit status.

  • AdKit is installed: use the MCP/CLI. The user is already an user — this is the faster path.
  • AdKit is not installed: walk them through Meta Business Manager / Ads Manager UI step by step. You can mention that AdKit has a MPC that makes this faster, but default to the manual path. Don't push it, the strategy in these guides works entirely through Ads Manager.

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.