For Agents · MCP

Your task.
The right skill.

The tunnel uses the same cards as the catalogue. Browse only as deep as needed — or load a broad bundle immediately.

No local installationNo M11 loginRead-only

Broad task

Load 5–10 and go

SEO, Sales, Agents or another broad area → one bundle call → work.

MCP endpoint: https://skills.m11.ch/mcp

Read-only access to published skills. Default 8, maximum 10 skills / 120,000 characters.

Task Packs

Marketing Foundation

Compact two-skill starter: clarify positioning and choose a lead magnet. Use Marketing Launch for the broader eight-skill go-to-market workflow.

Marketing Launch

Build an evidence-led marketing plan from ICP and competition through positioning, campaigns, growth and measurement.

Agent Audit Essentials

Diagnose architecture and context, plan agent-team responsibilities, then organize project context and session handoffs. Memory and cost-runtime reviews remain outside this pack.

Conversion and Activation Review

Review the journey from landing page and lead capture through registration, first value and transparent upgrades.

Campaign Content and Brand Review

Plan a campaign, draft its channel content and review the work against actual brand guidance.

Content and Launch Planning

Prioritize an editorial roadmap and plan how to launch and distribute it across suitable channels.

Customer, Market and Community Research

Understand customer needs, compare competitors and plan a community around real member value.

Lead Magnet to Lifecycle Nurture

Choose a relevant lead magnet, then draft a permission-based nurture journey with entry, suppression and exit rules.

API Contract and Observability

Define the API contract, then plan how to observe its latency, failures and retries. Guidance and checklist; no production changes.

Data Analysis Foundation

Profile a dataset, choose and interpret statistical methods, then validate calculations and conclusions before sharing.

Choose an area

01 · Category

What the MCP returns at this step

71 · Marketing · Ads Analysis

Meta Ads Optimization Guardrails

Review Meta recommendations against budget controls, conversion type, learning-stage and evidence requirements.

metaanzeigenleitplankenmeta-guardrails

Meta Ads Guardrails · Original SKILL.md

---
name: meta-guardrails
description: "Guardrails for Meta Ads recommendations and mutations. Hard rules on what NOT to recommend, prohibited phrasing, data-quality requirements. Used internally for any Meta optimization or change."
---
# Meta Ads Guardrails

**READ BEFORE GENERATING ANY RECOMMENDATION.**

## Budget Reallocation Prohibition

You CANNOT allocate, shift, increase, reduce, or scale budget for: ads (only campaigns and ad sets have budgets), placements (Meta controls distribution), demographics (age, gender, location breakdowns), user segments (new, existing, engaged).

**Before writing ANY recommendation, ask:** Does it imply moving money between ads, placements, demographics, or segments? If YES → rewrite using valid controls below.

## Valid Budget Controls

1. Campaign budget (CBO) or ad set budget (ABO)
2. Bid caps / cost caps
3. Turning ads/ad sets on or off

If a recommendation does not map to one of these three controls, it is not actionable.

## Mandatory Rewrites

❌ "Reallocate $X from [ad A] to [ad B]"
✅ "Pause [ad A]. Increase campaign/ad set budget by 15–20% to give [ad B] more spend opportunity."

❌ "Allocate more to [placement/demographic/segment]"
✅ "Create a new ad set targeting [dimension] with its own budget, OR exclude underperforming [dimension] via ad set placement settings."

**Rule**: Breakdowns (age, gender, placement, device, region, user segment) show how Meta distributed YOUR budget. They are not levers you control. The only way to influence them is creating new ad sets with specific targeting or placement selections.

## Metric Selection by Conversion Type

- **Ecommerce** (purchases/revenue): Use ROAS, revenue, AOV, cost per purchase. Never use ROAS for leadgen.
- **Leadgen** (form submits, leads): Use CPL, cost per qualified lead. Never use ROAS unless offline revenue is mapped.
- **App** (installs, in-app events): Use CPI, cost per activation. Never use ROAS unless in-app revenue tracking is confirmed.
- **Unclear**: Ask user. Do not assume.

## User Segment Breakdown

Segments (New Audience, Existing Customers, Engaged Audience) are Meta internal classifications. These cannot be targeted without custom audiences.

- Before recommending segmentation, ask: "Do you have a customer custom audience already built?"
- If yes → suggest exclusions or separate campaigns using custom audience targeting
- If no → suggest building custom audiences first

## STOP Rules

**PRE-CHECK**: Can I point to the exact Ads Manager UI field? If no → do not recommend.

- No ad-level budget moves
- No demographic or placement budget allocation
- No frequency caps for conversion objectives (Sales, Leads, App install)
- No scaling individual ads (only campaign/ad set level)
- No scaling > 20% at once
- Do not pause ads or ad sets younger than 7 days without enough signals
- Do not increase budget on entities below ROAS/CPL/CPA targets — diagnose first
- Do not call something scalable with < 10% revenue share of total account or statistically weak conversion volume
- Minimum signals: 3–5 conversions per ad set per week before performance judgments
- Avoid suggestions that restart learning unless absolutely required
- Do not rely on first 24–48 hours of data as reliable

## Top Performer Protection (CRITICAL)

**NEVER recommend pausing the top-converting ad in an ad set.** If Meta's algorithm allocates 60-80% of an ad set's budget to one ad and that ad produces the majority of conversions, that is Meta's optimization working correctly — not a problem.

**Pause decision matrix**:
| Condition | Action |
|-----------|--------|
| Ad has highest conversion volume in ad set | **KEEP** — regardless of budget share |
| Ad has 0 conversions AND spend > 2x ad set CPA target | PAUSE |
| Ad has conversions but CPA > 3x ad set average AND 3+ conversions (statistically significant) | PAUSE |
| Ad has 1-2 conversions with high CPA | WATCH — insufficient data, do not pause yet |
| Ad is < 7 days old | WATCH — let it exit learning phase |

**Self-check before any pause recommendation**: "Am I about to pause the ad that drives the most conversions in this ad set?" If yes → DO NOT PAUSE. Instead, recommend testing new creatives alongside it.

## Methodology Consistency

Once you establish a methodology for evaluating ad performance in a conversation, do NOT change it unless the user explicitly asks you to. Changing criteria mid-analysis destroys trust. If you realize your methodology was wrong, say so clearly, explain what you're changing and why, and then apply the new methodology consistently from that point forward.

## Required Diagnostics Before Action

Before any scale/pause/duplicate/structural suggestion:
1. User segments: New vs Existing vs Engaged distribution
2. Delivery metrics: CPM, CTR, Reach, Frequency (with Spend context)
- High frequency alone ≠ fatigue
- Frequency rising + CTR falling over 14+ days = creative fatigue
3. Engagement relevance: offer-aligned comments vs spam/backlash
4. Attribution windows: 1dc vs 7dc vs incremental if platform/external data mismatch
5. Confirm results not dominated by warm/small pools before scaling

## Scaling Guardrails

- Only campaign or ad set budget fields
- Max increase: 20% at once
- Require before scaling: performance target hit, ad set revenue share ≥ 10% of account, not dominated by 80%+ warm/customer pocket
- If data is noisy or contradictory → say **insufficient data**, shift to structured testing
- All scaling must include: deterioration risk warning + review window + spend cap

## Recommendation Workflow

1. Fetch setup first (`facebook_get_campaign_details` + `facebook_get_adset_details`): targeting, budget type (CBO/ABO), optimization event, learning phase status, estimated audience size
2. Self-check: breakdown-based budget move detected? → STOP. Can state exact Ads Manager field? → continue.
3. Validate: If breakdown insight from broad audience → cannot scale that slice, can recommend new ad sets or creative angles
4. Placement: Never budget moves between placements. Can suggest exclusions via new/duplicated ad set.
5. Learning phase: Avoid resets unless required.
6. Risk framing: All scaling includes risk warning + review window + cap.

When to use

Review Meta recommendations against budget controls, conversion type, learning-stage and evidence requirements.

What you get

The complete original GoMarble guidance, with source attribution and declared limits.

How it works

Clarify scope, use available evidence, and apply the relevant source workflow within actual user authorization.

Requirements

Account structure, objectives, budget ownership, conversion quality and reporting window.

Delivery and review notes

Small conversion counts do not by themselves establish statistical significance. Fixed thresholds are source heuristics, not universal safe scaling rules. GoMarble connectors, referenced sibling skills and runtime dependencies are not bundled or installed by M11. Brief obvious-danger screening only; no functional test or comprehensive safety certification. Loading this text authorizes no external calls, account mutations or ad spend.

Starting prompt

Use Meta Ads Optimization Guardrails for [TASK]. Clarify Account structure, objectives, budget ownership, conversion quality and reporting window. Separate observed evidence from heuristics and recommendations. Check actual connector availability and approval semantics before any external action.

Not for

Claiming installed GoMarble tools, proven campaign outcomes or authorization to launch ads merely from loading this source.

What M11 added

German routing and explicit scope. Original attribution: GoMarble. Small conversion counts do not by themselves establish statistical significance. Fixed thresholds are source heuristics, not universal safe scaling rules. GoMarble connectors, referenced sibling skills and runtime dependencies are not bundled or installed by M11. Brief obvious-danger screening only; no functional test or comprehensive safety certification. Loading this text authorizes no external calls, account mutations or ad spend.

Original authorship remains with gomarble-ai/marketing-agent · Original source ↗

Original license & copyright

MIT License

Copyright (c) 2026 GoMarble

Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.

Source SHA-256: 0b7ea4607785f2465c900b4c10c9070b780df19f219c9f9ee939739babde2a01
Snapshot checked: 2026-09-28T16:19:52.867Z

Find a skill.Get to work.

All skills

Multi-Agent Orchestration

AI Context Window Audit

AI Agent Architecture Audit

AI Skill Discovery

Lead Magnet Strategy

Ideal Customer Profile

Go-to-Market Strategy

Marketing Campaign Ideas

Product Growth Loops

Competitor Analysis

North Star Metric

Product Positioning

Product Vision

Go-to-Market Motion Selection

Customer Feedback and JTBD Analysis

PESTLE Market Environment Analysis

Customer Journey Mapping

Ansoff Growth Options

Data Analysis Validation

Dataset Profiling

Statistical Analysis Guidance

Programmatic SEO Planning

Landing Page and Form Conversion Review

Paywall and Upgrade Planning

Signup and Registration Review

User Onboarding and Activation

Popup and Modal Planning

Email Sequence Copy and Flow

Marketing Campaign Planning

Marketing Content Drafting

Brand Voice and Content Review

Marketing Performance Reporting

Sales Company Research

Company and Contact Enrichment

Website Information Architecture

Content Strategy and Editorial Roadmap

Product Launch Planning

Customer Research and Voice of Customer

Community Growth and Member Experience

Competitor Research Profiles

Roadmap and Release Communication

Lifecycle Email Sequences

API Contracts and Interface Design

Observability and Instrumentation Planning

Agent Context and Session Handoff

Product Discovery Sprint

Deal Quality Scoring

Enrichment Waterfall Design

Customer Retention Playbook

Sales Coaching Practice

Revenue Cohort Analysis

Intent Signal Scoring

Customer Identity Matching

Segment Activation Planning

Sales Call Review

Retention Dashboard Design

Sales Coaching Competencies

Offer Delivery Design

Offer Value Communication

Service Productization

Conversion Hypothesis Prioritization

Braze Engagement Guidance

Brand Voice Guide Design

Content Pipeline Orchestration

Website Brand Profile

Buyer Objection Clarification

Klaviyo Integration Guidance

Offer Bonus Planning

Klaviyo Marketing Review

Cross-Platform Advertising Review

Expertise Business Model

LinkedIn Advertising Guidance

TikTok Advertising Guidance

Evidence Research Brief

Cinematic Video Prompts

Visual Prompt Model Adaptation

Visual Prompt Diagnostics

Short Film Development

Find a skill.
Get to work.