M.A.D. SEO CONTENT

Autonomously plans, drafts, audits, and optimizes SEO content with omnichannel research, GEO-targeting, internal linking, schema, and analytics integration.

Security Scan

The skill's requirements, files, and runtime instructions are internally consistent with an autonomous SEO content manager; it asks to read/write only within the shared project workspace and delegates credentialed API access to an api-gateway skill.

Purpose & Capability

Name/description (autonomous SEO content + analytics) matches the declared permissions, dependencies, and instructions. Declared dependencies (api-gateway, scrapling-official, agent-browser) are referenced in SKILL.md and index.js for analytics, social discovery, and live search — these align with the stated capabilities. fs read/write permissions are limited to /root/.openclaw/shared/, which is appropriate for storing project files and the SQLite calendar.

Instruction Scope

Runtime instructions focus on research, drafting, auditing, schema generation, internal-link injection and analytics. They explicitly instruct the agent to operate only in designated shared project directories and to present diffs + require confirmation before editing existing files. Two items to note: (1) SKILL.md and index.js describe 'bypassing AI detectors with statistical burstiness' (ethical/abuse concern, not a direct security incoherence), and (2) onboarding asks the user for GA4/GSC property URLs and sitemap URLs — users should avoid pasting sensitive tokens/URLs that embed auth. Otherwise the instruction scope stays within the stated purpose and declared permissions.

Install Mechanism

This is instruction-and-code only (no external install spec, no downloads). The code provided is a JS tool wrapper that returns instructions; there are no remote installers or archived downloads referenced. Low install-surface risk.

Credentials

The skill declares no required env vars and does not request credentials directly. It relies on an api-gateway dependency to handle GA4/GSC credentials, which is consistent and proportional. The only filesystem access requested is read/write to the shared workspace (declared in skill.json).

Persistence & Privilege

always:false (not forced). The skill requests standard read/write access to its own shared workspace only and does not attempt to modify other skills or system-wide settings. It also documents safety controls (dry_run, mandatory confirmation for mutations).

Assessment

This skill is coherent with its SEO purpose, but review a few practical points before installing: 1) It will write/read files under /root/.openclaw/shared/ (PROJECT_STRATEGY.json, ENTITY_MAP.json, mad_seo.db). Ensure you are comfortable with that location and inspect these files after use. 2) The skill asks you to provide GA4/GSC property URLs and sitemap URLs during onboarding — do not paste API tokens or credential-bearing URLs into the chat; instead rely on the api-gateway skill to hold credentials. 3) The skill depends on third-party skills (api-gateway, scrapling-official, agent-browser) — verify you trust those skills because they will perform networked operations on the skill's behalf. 4) Use dry_run modes and require the visual diff confirmation when running inject_internal_links or other mutating tools. 5) Note the author claims the tool will help 'bypass AI detectors'—that is an ethical/abuse concern (content-policy risk) rather than a technical incoherence; consider whether that behavior aligns with your policies before using. If you want higher assurance, review the api-gateway and scrapling-official skills' implementations and run initial operations in dry_run mode.

M.A.D. SEO CONTENT (Multimodal Attribution & Discovery) is the ultimate autonomous content and authority engine for OpenClaw. Built for the 2026 SEO/GEO Landscape, it transforms the standard SEO writing process into a director-level management workflow oriented toward Omnichannel Authority, Multimodal Attribution, and High-Quality Content Synthesis.

Version 1.1.0 introduces the Lean 11-Tool Architecture, merging overlapping tools to ensure lightning-fast context efficiency, zero hallucination rates, and flawless autonomous execution.

---

🛠️ The Lean 11-Tool Suite

1. mad_seo:onboard

Triggers the interactive onboarding workflow to verify dependencies and securely configure your Target Market, Brand Name, and Project Paths. Trigger: "Onboard me to M.A.D. SEO CONTENT"

2. mad_seo:research_strategy

Performs Omnichannel 3.0 Analysis. It crawls SERP Top-10 AND community sentiment from platforms like Reddit to find unique "experience-based" insights. Trigger: "Omnichannel research for [Topic]"

3. mad_seo:analyze_share_of_voice [RESTORED]

Analyzes AI search responses (Gemini, Perplexity) to calculate your brand's AI Citation Score. Trigger: "Analyze share of voice"

4. mad_seo:plan_strategy [MERGED]

Replaces generate_roadmap & plan_content. Generates the 12-month strategic roadmap and specifically creates the next 4 weeks of titles, classifying them by Funnel Stage (TOFU/MOFU/BOFU) and saving them to the project database. Trigger: "Plan my strategy"

5. mad_seo:draft_article [MERGED]

Replaces draft_article, humanize_content & headline_pro. A single-pass powerhouse. The agent drafts a 1,500-3,000 word article, natively applies the Human-GEO Framework (bypassing AI detectors with statistical burstiness), and generates 3 high-CTR headline options at the top of the file. Trigger: "Draft an article"

6. mad_seo:audit_eeat [MERGED]

Replaces audit_eeat & audit_author. Performs a unified Scientific Audit. It checks the article for "Information Gain" (ensuring 3 unique concepts not found in Top 10), flags hallucinations, AND evaluates the Author's bio to secure "Expertise" markers. Trigger: "Audit eeat"

7. mad_seo:generate_schema

Generates GEO-optimized JSON-LD (Article, FAQ, HowTo) with Knowledge Graph relationship anchoring (Wikidata). Trigger: "Generate schema"

8. mad_seo:site_wide_intelligence

Deep audit linking Sitemap URLs to GSC performance. It builds a Global Entity Map for optimal internal linking architecture. Trigger: "Run site intelligence"

9. mad_seo:inject_internal_links [SAFE]

Finds existing, high-authority markdown posts and proposes contextual links to a newly published article.

• Dry-run Mode: Set dry_run: true to see proposed changes without editing.
• Confirmation Prompt: The agent will ALWAYS show a diff and wait for your approval before modifying any existing file. Trigger: "Inject internal links"

10. mad_seo:repurpose_content

Executes the 'One-to-Many' workflow, slicing a pillar post into Twitter threads, LinkedIn posts, and Newsletter teasers. Trigger: "Repurpose content"

11. mad_seo:analytics_suite [MERGED]

Replaces performance_audit, content_decay & cannibalization. A unified analytics engine that talks to GA4/GSC via API. It handles 3 specific audit types:

• low_ctr: Finds pages with high impressions but no clicks.
• decay: Finds old pages losing traffic to schedule a refresh.
• cannibalization: Finds URLs fighting each other for the same keyword. Trigger: "Run analytics suite"

---

🧠 The Human-GEO Framework

Built directly into the drafting tool, it enhances content quality via:

1. Linguistic Variety: Mixes sentence lengths to ensure high-quality, human-like flow.
2. Vocabulary Refinement: Automatically replaces repetitive phrasing with precise, industry-specific terminology.
3. Clarity Optimization: Replaces hyperbolic 'fluff' with direct, evidence-based facts.
4. Contextual Anchoring: Injects specific community findings to ground the content in real-world expertise.

🔒 Security & Safety

M.A.D. SEO CONTENT is designed with a "Safety-First" autonomous architecture:

1. Explicit Whitelisting: The agent is strictly instructed to only interact with files in your designated project directories (e.g., ./shared/content/).
2. Review Before Mutate: The inject_internal_links tool requires a mandatory User Confirmation of a visual diff before any existing file is modified.
3. Dry-Run Mode: All high-privilege tools support a dry_run parameter to preview actions without committing changes.
4. Path Safety: The agent is restricted to its own project state and is instructed to ignore unrelated sensitive state or credentials in the environment.

🧩 Dependencies & Authentication

[!IMPORTANT] To unlock Version 1.1 features, the following skills are required:

1. api-gateway — Mandatory for Analytics. This skill manages all GA4/GSC credentials securely. M.A.D. SEO CONTENT never sees your raw API tokens; it only sends authorized requests via this gateway.
2. scrapling-official — Social discovery (Reddit/Forums).
3. agent-browser — Live AI search analysis.