seo-agi for Openclaw

An elite technical SEO agent that creates entity-rich, auditable content designed to rank on Google and be cited by generative AI engines.

gbessoni
v0.1.0
Mar 18, 2026
0
835
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install seo-agi

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install seo-agi using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is seo-agi?

SEO-AGI represents a paradigm shift in content creation within the Openclaw Skills ecosystem. It is specifically engineered for Generative Engine Optimization (GEO), moving beyond keyword stuffing to focus on information gain and semantic authority. The skill automates the production of high-fidelity landing pages, local service guides, and comparison tables that satisfy both traditional search algorithms and modern LLM retrieval systems like Perplexity and ChatGPT.

By prioritizing real operational data over generic AI-generated fluff, this skill ensures that every page passes the Reddit Test—meeting the standards of knowledgeable practitioners. It integrates deeply with competitive intelligence tools to identify gaps in existing search results, allowing developers to build pages that offer genuine value and unique insights that competitors miss.

seo-agi Use Cases

  • Creating airport parking and local service pages with high conversion intent.
  • Optimizing existing web content to secure citations in AI search summaries.
  • Conducting automated SERP and keyword research using premium data sources.
  • Generating technical comparison tables and pricing guides with embedded schema.
  • Auditing content for EEAT signals and information gain to improve ranking resilience.

How seo-agi Works

  1. The agent locates the skill root and initializes competitive intelligence scripts via the command line.
  2. It performs deep research using DataForSEO, Ahrefs, or SEMRush to analyze the top 10 organic results.
  3. A comprehensive page brief is constructed, identifying search intent, competitive gaps, and target word counts.
  4. Content is authored using a 500-token chunk architecture, ensuring high retrievability for AI answer engines.
  5. The agent inserts mandatory tables, "Not For You" blocks, and specific operational facts to satisfy quality filters.
  6. Semantic markup is embedded using RDFa and Microdata to label entities for search engine crawlers.
  7. A final verification audit is performed, tagging all claims with auditable source markers before delivery.

seo-agi Setup

To deploy this skill within your Openclaw Skills workflow, install the necessary dependencies and configure your environment variables.

# Install core dependencies
pip install requests

# Optional: Install GSC integration tools
pip install google-auth google-api-python-client

Configure your API keys in ~/.config/seo-agi/.env to enable live research:

DATAFORSEO_LOGIN=your_credentials
DATAFORSEO_PASSWORD=your_password
GSC_SERVICE_ACCOUNT_PATH=/path/to/service-account.json

seo-agi Data Schema & Taxonomy

SEO-AGI maintains a structured data layer to ensure all generated content is backed by verifiable research. The following table describes the organizational structure used by this skill:

Component Format Storage Location
Competitive Briefs JSON ~/.local/share/seo-agi/research/
Generated Pages Markdown ~/Documents/SEO-AGI/pages/
Page Rewrites Markdown ~/Documents/SEO-AGI/rewrites/
Schema Templates YAML/MD ${SKILL_ROOT}/references/
Audit Logs Log ${SKILL_ROOT}/logs/

seo-agi Advanced Features

  • Integrated MCP tools for direct connectivity with Ahrefs and SEMRush keyword databases.
  • Automated cannibalization detection by cross-referencing Google Search Console data.
  • Advanced 500-token chunking strategy optimized for modern transformer-based retrieval windows.
  • Self-correcting quality filters that reject generic "AI slop" in favor of hard operational data.
  • Multi-agent batch processing for generating entire location-based content hubs simultaneously.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*