How do I install the Agentic Marketing Intelligence skill?

Run npx skillmds@latest add cleanexpo/agentic-marketing-intelligence in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Agentic Marketing Intelligence safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Agentic Marketing Intelligence?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Agentic Marketing Intelligence free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Agentic Marketing Intelligence?

cleanexpo (@cleanexpo) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Agentic Marketing Intelligence by cleanexpo

Name: Agentic Marketing Intelligence
Author: cleanexpo

Agentic Marketing Intelligence

Research → verify → score → plan → improve loop for SEO / AEO / GEO across the Unite-Group portfolio. Ingests Obsidian source data and (when supplied) YouTube channels, cross-verifies every claim against ≥4 sources, scores ranking opportunity / decay / freshness / GEO visibility with confidence-adjusted prioritisation, and produces human-gated site-improvement tickets. Never fabricates metrics. Orchestrates — does not duplicate — the existing google-search-console, google-business-profile, local-seo-agent, google-updates-sentinel, competitive-local-strategy and algorithm-knowledge-base skills.

cleanexpo Updated Sep 17, 2026

File contents

Agentic Marketing Intelligence 2026

Prime directive

Continuously improve portfolio websites using real evidence, real data, and measurable search-performance signals — never invented metrics, never opinion treated as fact.

Non-negotiable hard rules

No fabricated data. No invented rankings, impressions, CTR, volume, traffic, or conversions. Missing data is tagged DATA_REQUIRED. Unverifiable claims are UNVERIFIED_CLAIM. Opinion is OPINION_SOURCE. Promising-but-unproven is HYPOTHESIS_FOR_TESTING.
Every recommendation ties to: a source · a reason · a measurable signal · a proposed action · a validation method. (Enforced by quality-gates.md.)
Confidence gates impact. The master prioritiser is (impact · confidence) / risk. Influencer advice (OPINION_SOURCE = 0.1) can never out-rank a CONFIRMED action of similar impact.
No live publishing without a human gate. This skill prepares; humans approve. See ../../docs/marketing-intelligence/human-approval-gates.md.
Orchestrate, don't duplicate. Local-pack/GBP/GSC mechanics belong to the existing skills; this skill consumes their output and adds cross-portfolio scoring + prioritisation.
Internal first-party data overrides generic advice whenever they conflict.

What it does (one line each)

Ingest Obsidian Sources/ + supplied YouTube channels → source map.
Extract claims/tactics/frameworks → claim dataset.
Verify each claim against ≥4 references → confidence-rated ledger.
Score pages on the 12 models in scoring-models.ts.
Prioritise by confidence-adjusted action score; block high-risk / placeholder items.
Plan human-gated tickets with rollback notes.
Loop on daily/weekly/monthly cadences (../../docs/marketing-intelligence/automation-schedule.md).

Files in this skill

File	Purpose
`skill.md`	This contract
`README.md`	Quick start + how it fits the ecosystem
`workflow.md`	The 7-agent run, step by step
`agent-prompts.md`	The prompt for each of the 7 agents
`quality-gates.md`	The 6 gates a run must pass
`scoring-models.ts`	The 12 formulas (pure, tested-shape, type-checked)
`types.ts`	Shared types incl. `ConfidenceLabel`, `DataStatus`, `Weights`
`inputs.schema.json` / `outputs.schema.json`	Run contract
`source-map.schema.json` / `claim-verification.schema.json`	Artifact contracts
`examples.md`	Worked example with placeholder→real transition

First-run status (2026-05-29)

Run #1 output is in ../../docs/marketing-intelligence/. YouTube + live-metric deliverables are DATA_REQUIRED because the vault contained no channel references and no GSC/Semrush exports were present. The skill, schemas, math models, claim ledger, design docs and CEO synthesis are complete.

---name: agentic-marketing-intelligencedescription: Research → verify → score → plan → improve loop for SEO / AEO / GEO across the Unite-Group portfolio. Ingests Obsidian source data and (when supplied) YouTube channels, cross-verifies every claim against ≥4 sources, scores ranking opportunity / decay / freshness / GEO visibility with confidence-adjusted prioritisation, and produces human-gated site-improvement tickets. Never fabricates metrics. Orchestrates — does not duplicate — the existing google-search-console, google-business-profile, local-seo-agent, google-updates-sentinel, competitive-local-strategy and algorithm-knowledge-base skills.---# Agentic Marketing Intelligence 2026## Prime directiveContinuously improve portfolio websites using **real evidence, real data, and measurablesearch-performance signals** — never invented metrics, never opinion treated as fact.## Non-negotiable hard rules1. **No fabricated data.** No invented rankings, impressions, CTR, volume, traffic, or conversions. Missing data is tagged `DATA_REQUIRED`. Unverifiable claims are `UNVERIFIED_CLAIM`. Opinion is `OPINION_SOURCE`. Promising-but-unproven is `HYPOTHESIS_FOR_TESTING`.2. **Every recommendation ties to:** a source · a reason · a measurable signal · a proposed action · a validation method. (Enforced by `quality-gates.md`.)3. **Confidence gates impact.** The master prioritiser is `(impact · confidence) / risk`. Influencer advice (`OPINION_SOURCE` = 0.1) can never out-rank a `CONFIRMED` action of similar impact.4. **No live publishing without a human gate.** This skill prepares; humans approve. See `../../docs/marketing-intelligence/human-approval-gates.md`.5. **Orchestrate, don't duplicate.** Local-pack/GBP/GSC mechanics belong to the existing skills; this skill consumes their output and adds cross-portfolio scoring + prioritisation.6. **Internal first-party data overrides generic advice** whenever they conflict.## What it does (one line each)- **Ingest** Obsidian `Sources/` + supplied YouTube channels → source map.- **Extract** claims/tactics/frameworks → claim dataset.- **Verify** each claim against ≥4 references → confidence-rated ledger.- **Score** pages on the 12 models in `scoring-models.ts`.- **Prioritise** by confidence-adjusted action score; block high-risk / placeholder items.- **Plan** human-gated tickets with rollback notes.- **Loop** on daily/weekly/monthly cadences (`../../docs/marketing-intelligence/automation-schedule.md`).## Files in this skill| File | Purpose || ----------------------------------------------------------- | ------------------------------------------------------------- || `skill.md` | This contract || `README.md` | Quick start + how it fits the ecosystem || `workflow.md` | The 7-agent run, step by step || `agent-prompts.md` | The prompt for each of the 7 agents || `quality-gates.md` | The 6 gates a run must pass || `scoring-models.ts` | The 12 formulas (pure, tested-shape, type-checked) || `types.ts` | Shared types incl. `ConfidenceLabel`, `DataStatus`, `Weights` || `inputs.schema.json` / `outputs.schema.json` | Run contract || `source-map.schema.json` / `claim-verification.schema.json` | Artifact contracts || `examples.md` | Worked example with placeholder→real transition |## First-run status (2026-05-29)Run #1 output is in `../../docs/marketing-intelligence/`. YouTube + live-metric deliverables are`DATA_REQUIRED` because the vault contained no channel references and no GSC/Semrush exports werepresent. The skill, schemas, math models, claim ledger, design docs and CEO synthesis are complete.

Frequently asked questions