Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsCommunityBlog
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

Back to skills

Launch Meta Ad Campaign

ASecurity

Plan or safely create Meta/Facebook/Instagram campaigns through GooseWorks tools or a bundled direct Meta Marketing API adapter. Use automatically for requests to launch, run, create, publish, set up, or plan Meta ads; use reporting skills for analysis-only requests.

1,220 stars
0 votes
0 copies
0 views
Added 9/22/2026
ai-agentsgoapi

Works with

cliapi

Security Analysis

A100/100

Scanned 9/22/2026

Install to Claude Code

$npx -y skills add gooseworks-ai/goose-skills --skill launch-meta-ad-campaign --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Launch Meta Ad Campaign?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Launch Meta Ad Campaign
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/gooseworks-ai-launch-meta-ad-campaign/badge)](https://www.skillsdirectory.com/skills/gooseworks-ai-launch-meta-ad-campaign)

More formats (shields.io, HTML) on the badges page.

Download with Pro
Files
SKILL.md
---
name: launch-meta-ad-campaign
description: Plan or safely create Meta/Facebook/Instagram campaigns through GooseWorks tools or a bundled direct Meta Marketing API adapter. Use automatically for requests to launch, run, create, publish, set up, or plan Meta ads; use reporting skills for analysis-only requests.
---

# Launch Meta Ad Campaign

Guide the user from campaign intent to either a verified **paused** Meta campaign or a complete campaign plan. This skill is agent-neutral: choose the best available adapter, but keep one safety workflow and one approval boundary.

## Choose one mode

Use the first available mode:

1. **GooseWorks adapter** when all GooseWorks launch tools are callable. Read [references/gooseworks-adapter.md](references/gooseworks-adapter.md).
2. **Direct Meta adapter** when a valid Meta token and ad-account access are available. Read [references/direct-meta-adapter.md](references/direct-meta-adapter.md) and use the bundled `tools/meta_marketing_api.js`.
3. **Planning-only fallback** when neither write adapter is available. Read [references/planning-only.md](references/planning-only.md). Produce a useful plan; do not present missing write access as a failed planning task.

If both write adapters are available, prefer GooseWorks because its approval page, creative links, and recovery records are already integrated. Never mix adapters within one launch.

## Universal workflow

### 1. Resume or inspect before asking

- Resume any existing launch record or push first. Never duplicate a partially completed launch.
- Inspect existing campaigns, ad sets, ads, and relevant creative before recommending a new structure.
- Explain whether the intent fits reuse or needs a new campaign. Do not create a second context system when the selected adapter already provides one.

### 2. Collect only missing decisions

State what is known, then ask once for the missing budget, account-local schedule, destination URL, result/objective, bidding and optimization, geography/audience, placements, tracking, Page/Instagram identity, and creative/copy.

Do not silently approximate unsupported objectives, special-ad-category handling, targeting, bid controls, or tracking. Planning-only mode may still design these configurations even when the available write adapter cannot create them.

### 3. Validate before any write

- Render or inspect the final destination, follow redirects, require public HTTPS, and confirm the approved host.
- Check message and claim match, one clear CTA, tracking/pixel availability, and obvious broken-page states.
- Run Meta policy review for the objective, audience, copy, creative, destination, and special-category implications.
- Block hard destination or policy failures. Show non-blocking warnings plainly and explain any optimization fallback.

### 4. Prepare one exact plan

Summarize in human language:

- campaign name, objective, special category, and paused status;
- ad set names, budget type and amount, account timezone, dates, geography, audience, placements, optimization, and bidding;
- every ad, creative, Page/Instagram identity, copy, CTA, and destination;
- what is reused versus created, destination/policy results, warnings, and tracking assumptions.

Persist the exact plan before asking for approval. A changed plan requires validation and approval again.

### 5. Obtain explicit human approval immediately before writing

Preparation, inspection, validation, and planning are read-only. They do not authorize a Meta write.

- GooseWorks mode uses the returned approval page/card. Chat cannot approve a GooseWorks push.
- Direct Meta mode requires an explicit user confirmation for the exact paused plan immediately before the publish command.
- A decline or missing confirmation means **no write**.

Never activate ads automatically. This skill has no direct-activation step.

### 6. Create paused, persist, and read back

- Create campaign, ad set, and ads only with `PAUSED` status.
- Persist each returned ID immediately so interruption and partial failure are recoverable.
- Read every object back from Meta. Report success only after names, parent links, objective, budget/schedule, targeting/optimization, creative/destination identity, and paused state match the approved plan.
- On failure, list what failed, every object already created, its verified status, and the exact next action. Resume from persisted IDs only after renewed user approval; never recreate completed steps.

## Final response

End with the adapter used, inspection/reuse decision, approved configuration, destination and policy result, Meta IDs, readback status, warnings or partial failures, and the next human action. The normal write-mode stopping point is: **ready in Meta, paused, spending nothing**.

Attribution

gooseworks-aigooseworks-ai
View sourceMore from gooseworks-ai →
SSkills DirectorySkills Directory

Your tool, in front of Claude Code builders.

3 founder slots · $299/mo · GSC-verified traffic · sponsors can never buy grades.

See placements

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Your tool, in front of Claude Code builders.

3 founder slots · $299/mo · GSC-verified traffic · sponsors can never buy grades.

See placements

Related Skills

Caveman

Ultra-compressed communication mode that cuts output tokens while keeping technical accuracy. Levels: lite, full, ultra and the wenyan variants. Use for /caveman, "caveman mode", "talk like caveman", "be brief" or "less tokens".

1066601 votes

Hyperplan

Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...

686011 votes

Mcp Code Execution

Routes multi-tool workflows through MCP servers for large datasets and pipelines. Use when Bash tool overhead is limiting throughput on data-heavy tasks.

3351 votes

catchup

Recovers the conversation and failed tool calls of a previous Codex, Claude Code, Antigravity, Cline, Copilot CLI, Cursor, DeepSeek Harness, Kimi, OpenCode, Pi Agent, or ZCode session. Use when the user says "catch up", "what did the last session do", "get me up to speed", "I switched agents", asks to recover/summarize a previous session before continuing, or asks to diagnose or report a catchup failure. Do NOT use for the current conversation, git history, or any non-agent log.

651 votes

math-skill

A comprehensive mathematical reasoning skill for AI assistants — handles arithmetic to research-level problems with rigorous step-by-step reasoning, systematic verification, and transparent uncertainty handling

381 votes
View all in ai-agents →