Systems Lab

Agent skill

ad-campaign-management

Use when creating, managing, or optimizing paid advertising campaigns across any platform.

slowingNeeds a keyActs undeclared1,295 words

Filed under ABM and paid.

From manojbajaj95/claude-gtm-plugin · 51 skills · 92 · pushed 2026-05-18

What it does when it runs

Use when creating, managing, or optimizing paid advertising campaigns across any platform. Covers ad copy generation, ad creative strategy, campaign management, and competitive intelligence. Triggers on: ad copy, ad creative, paid ads, paid advertising, Facebook ads, LinkedIn ads, Google ads, Meta ads, ad campaign, ad strategy, ad testing, competitive ads, competitor ads, ad library, ad analysis, ad creative generation, campaign management, ROAS, CPC, CTR. Use this skill whenever the user mentions ads, paid channels, ad campaigns, ad copy, ad performance, or wants to analyze competitor advertising.

Read from the skill and the 2 files bundled beside it. A skill’s own description is written to be selected by an agent, so it describes the job and not the dependencies.

Keys and connectors you must supply
  • ELEVENLABS_API_KEY
  • GEMINI_API_KEY
Hosts it reaches
  • ai.google.dev
  • api.elevenlabs.io
  • aws.amazon.com
  • cloud.google.com
  • developer.ideogram.ai
  • docs.bfl.ml
  • docs.cartesia.ai
  • docs.dev.runwayml.com
  • elevenlabs.io
  • fish.audio
  • generativelanguage.googleapis.com
  • higgsfield.ai
  • ideogram.ai
  • klingai.com
  • murf.ai
  • platform.openai.com
  • play.ht
  • replicate.com
  • seed.bytedance.com
  • voicebox.sh
  • www.remotion.dev
  • www.resemble.ai
  • www.wellsaid.io
Tool permissions it declares
No allowed-tools in the frontmatter. It does act, so it runs under whatever permissions your session already grants.
Actions present in the files
shellwrites files

Ask about ad-campaign-management

Opens your assistant with this page's verified links already in the prompt.

Is this safe to install?ClaudeChatGPT
Adapt it to my stackClaudeChatGPT
What else do I need for it to workClaudeChatGPT
Rather ask a human? Talk to Cheetah
git clone --depth 1 --filter=blob:none --sparse https://github.com/manojbajaj95/claude-gtm-plugin.git /tmp/claude-gtm-plugin
git -C /tmp/claude-gtm-plugin sparse-checkout set "skills/ad-campaign-management"
mkdir -p ~/.claude/skills/ad-campaign-management
cp -R "/tmp/claude-gtm-plugin/skills/ad-campaign-management/." ~/.claude/skills/ad-campaign-management/

Picked up without a restart. A project skill of the same name is shadowed by your personal one. For one repository only, swap ~/.claude/skills for .claude/skills. Claude Code docs ↗

Or take the whole library

This repo ships a .claude-plugin manifest, so Claude Code can install all 51 skills at once. Plugin skills are invoked as /<plugin>:<skill>, so they never collide with your own.

/plugin marketplace add manojbajaj95/claude-gtm-plugin
/plugin

The folder is the same in every client that implements the format — 46 of them — so if yours is not above, only the destination changes.

Before you install: this skill will not complete its job on a bare agent. It needs ELEVENLABS_API_KEY, GEMINI_API_KEY, which you have to obtain separately.

Reproduced in full from manojbajaj95/claude-gtm-plugin/blob/3c308fa4f7879d13d3d3df394247eabf6d26b684/skills/ad-campaign-management/SKILL.md, which is licensed MIT (repository). 1,295 words, 39 headings.

Ad Campaign Management

Workspace Context

Read bootstrap context before asking questions: strategy/brand.md for brand, audience, offer, channels, tools, constraints, and metrics; about/me.md for personal voice; content/ideas.md and content/calendar.md for content planning. Use legacy product-marketing context files only as fallback. Save generated drafts to content/<platform>/drafts/YYYY-MM-DD_short-topic-slug.md, and route durable learnings back to strategy/brand.md, about/me.md, or content/ideas.md.

Operating Contract

This skill is self-contained for its frontmatter scope: use its local instructions, references, scripts, and assets as the playbook; ask only for missing task-specific inputs; hand off to adjacent skills instead of expanding scope; and return an actionable artifact, decision, plan, draft, or diagnostic.

End-to-end paid advertising: write ad copy, develop creative strategy, manage campaigns, and analyze competitor ads.

Use the Workspace Context above first; only ask for ad-platform inputs not covered there.


1. Ad Copy Creation

Platform Constraints

PlatformElementLimit
Google Ads (RSA)Headline30 chars × 15
Google Ads (RSA)Description90 chars × 4
Meta (FB/IG)Primary text125 chars visible
Meta (FB/IG)Headline40 chars
LinkedInIntro text150 chars recommended
LinkedInHeadline70 chars recommended
TikTokAd text80 chars recommended
Twitter/XTweet280 chars

For full format specs, see references/platform-specs.md.

Copywriting Frameworks

AIDA — Attention → Interest → Desire → Action PAS — Problem → Agitate → Solution 4Ps — Promise → Picture → Proof → Push Before-After-Bridge — Current pain → Desired state → Your solution

Headlines That Work

  • Specific over vague: "Cut reporting time 75%" not "Save time"
  • Benefits over features: "Ship code faster" not "CI/CD pipeline"
  • Include numbers when possible: "3x faster," "in 5 minutes," "10,000+ teams"
  • Each RSA headline must stand alone — they're combined randomly
  • Include at least: 1 keyword headline, 1 benefit headline, 2–3 CTA headlines

Platform Best Practices

Google: Include keywords; use numbers/stats; power words (Free, Proven, New) Meta: Hook in first 3 words; ask questions; social proof performs well; emojis B2C only LinkedIn: Professional but not boring; lead with value; stats resonate with B2B TikTok: Casual/authentic tone; hook in 1–2 seconds; strong CTA at end

A/B Testing Variables

  • Headlines: Question vs. statement; benefit-led vs. feature-led; with/without numbers
  • Primary text: Short vs. long; problem-focused vs. solution-focused; with/without social proof
  • CTAs: Direct ("Buy Now") vs. soft ("Learn More"); benefit-based vs. action-based
  • Protocol: One variable at a time; min 7 days or 1,000 impressions; 95% significance before declaring winner

Output Format

Organize by angle with character counts:

## Angle: [Pain Point — Manual Reporting]

### Headlines (30 char max)
1. "Stop Building Reports by Hand" (29)
2. "Reports Done in 5 Min" (21)

### Descriptions (90 char max)
1. "Marketing teams save 10+ hours/week. Start free." (49)

Offer CSV format for 10+ variations:

headline_1,headline_2,description_1,platform
"Stop Manual Reporting","Automate in 5 Min","Save 10+ hrs/week. Start free.","google_ads"

2. Creative Strategy

Define Angles First

Before writing copy, establish 3–5 distinct angles (different reasons someone clicks):

CategoryExample
Pain point"Stop wasting time on X"
Outcome"Achieve Y in Z days"
Social proof"Join 10,000+ teams who..."
Curiosity"The secret top companies use"
Comparison"Unlike X, we do Y"
Identity"Built for [specific role]"
Contrarian"Why [common practice] doesn't work"

Generate Variations Per Angle

Vary: word choice, specificity (numbers vs. general), tone (direct/question/command), structure (short punch vs. full benefit).

Iterating from Performance Data

  1. Analyze winners — identify winning themes, structures, word patterns by CTR/CVR/ROAS
  2. Analyze losers — find what falls flat (too generic, wrong tone, too long)
  3. Generate new creative — double down on winning themes, test 1–2 new angles, avoid loser patterns
  4. Document iterations:
## Iteration Log
- Round: [N] | Date: [date]
- Top performers: [list with metrics]
- Winning patterns: [summary]
- New angles tested: [list]
- Angles retired: [list]

Generating Ad Visuals

For image and video creative, use AI tools. See references/generative-tools.md for:

  • Image generation: Flux, Ideogram, Nano Banana Pro (Gemini) for static ad images
  • Video generation: Veo, Kling, Runway, Sora, Seedance, Higgsfield
  • Voice/audio: ElevenLabs, OpenAI TTS for voiceovers
  • Code-based video: Remotion for templated, data-driven video at scale

Recommended workflow: Generate hero creative with AI → build Remotion templates from winners → batch-produce variations at scale.


3. Campaign Management

Key Metrics

TypeMetrics
CostCPM, CPC, CPL, CPA/CAC
PerformanceCTR, CVR, ROAS, LTV:CAC

Campaign Structure

Google Ads: Account → Campaign (budget, geo) → Ad Group (keyword theme) → Keywords + Ads

Meta Ads: Campaign (objective) → Ad Set (audience, placement, budget) → Ads (creative, copy)

Bidding strategies:

  • Target CPA — optimize for conversions at target cost
  • Target ROAS — optimize for conversion value
  • Maximize conversions — spend budget for max conversions
  • Manual — full control, time-intensive

Audience Targeting

Meta:

  • Core: demographics, interests, behaviors, location
  • Custom: website visitors, customer lists, engagement audiences
  • Lookalike: based on high-value customers, 1–10% similarity

LinkedIn:

  • Company: industry, size, name
  • Job: title, function, seniority
  • Professional: skills, groups, interests

Optimization Cadence

Daily: Budget pacing, cost metrics, major anomalies Weekly: CTR/CVR trends, top/bottom performers, search term review (Google) Monthly: Overall ROAS/ROI, audience performance, creative fatigue, strategic adjustments

Quick Optimization Wins

  1. Pause underperforming ads
  2. Increase bids on top performers
  3. Add negative keywords (Google)
  4. Refine audience targeting
  5. Refresh fatigued creative

Budget Allocation

  • Testing phase: 10–20% on new ideas
  • Proven performers: 60–70% of budget
  • Retargeting: 10–20% of budget
  • Scaling: increase budget gradually (max 20%/week) on winning audiences/creative

4. Competitive Intelligence

Use ad libraries (Facebook Ad Library, LinkedIn Ad Library) to extract and analyze competitor ads.

Extraction Process

Extract all current ads from [Competitor] on Facebook Ad Library
Scrape ads from [Company A], [Company B], [Company C] — compare approaches
Get LinkedIn ads from [Competitor] — analyze their B2B positioning

Analysis Output

For each competitor, deliver:

# [Competitor] Ad Analysis

## Overview
- Total ads: [N] active | Primary themes: [breakdown]
- Ad formats: [static/video split] | CTA patterns: [list]

## Key Problems They're Highlighting
1. [Pain point] ([N] ads) — Copy: "..." — Why it works: [reason]

## Successful Creative Patterns
- Pattern 1: [description + frequency]
- Pattern 2: [description + frequency]

## Best Headlines
1. "[headline]" → [why it works]

## Recommendations for Your Ads
1. [Actionable insight based on their patterns]

What to Analyze

  • Messaging: What problems they emphasize, how they position vs. competitors
  • Creative patterns: Visual styles, video vs. static, layout trends
  • Copy formulas: Headline structures, CTA patterns, length and tone
  • Campaign strategy: Seasonal campaigns, retargeting patterns, feature announcements

Competitive Intelligence Workflows

Ad Campaign Planning: Extract competitor ads → identify patterns → note messaging gaps → draft test variations

Positioning Research: Analyze 5 competitors → map their positioning → find underserved angles → develop differentiated messaging

Trend Tracking: Compare competitor ads across quarters to spot messaging evolution

Legal & Ethical

✓ Use for research and inspiration only ✓ Adapt patterns — don't copy ads directly ✓ Respect intellectual property ✗ Don't plagiarize copy or steal designs


5. References

Files bundled with it

These load only when the skill asks for them, so they cost nothing until it runs.

Other skills for the same job

Different authors, same problem. Matched on the words in the skill name, across every library in the catalogue except this one.

Need help setting it up?

This page tells you what ad-campaign-management does and what it needs. Cheetah builds the agent setup it runs inside: data, CRM, sequencing and the guardrails.

Book a call →

The directory stays free. There is nothing gated behind this.