An automated AI content imitation and SEO optimization skill that extracts high-authority sources, drafts brand-safe revisions, and exports publish-ready packages.
The fastest way to install a skill directly from the registry.
npx clawhub@latest install geo-deep-imitate
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
Copy this prompt to OpenClaw to install it automatically.
Help me install geo-deep-imitate using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Get the raw skill files in a ZIP archive.
GEO Deep Imitate is a sophisticated content adaptation and generation engine designed for the Openclaw ecosystem. Operating autonomously or under orchestrators like geo-cycle-autopilot or geo-opt-coordinator, this skill automates the extraction, purification, and rewriting of top-performing web sources. By utilizing advanced retrieval and deep imitation capabilities, it transforms noisy competitor URLs into brand-safe, highly optimized alternative drafts.
As a foundational component of modern Openclaw Skills, it eliminates manual editorial workflows. Content teams can reliably crawl raw source content, strip out distracting components like ads or navigation, generate precise drafts aligned with target outlines, and assemble local backup packages for mass deployment.
Ensure that your local execution environment has access to the built-in web fetch tools of your AI agent framework. Openclaw Skills require no heavy external browser scraper setups to operate this skill.
Create your local API key file using your credentials from SaaS Account Settings under Dragon Key:
mkdir -p ~/.qclaw
echo -n "your_geo_api_key_here" > ~/.qclaw/geo-api-key
Verify that your API key is configured correctly before starting any processing cycle:
curl -s -X POST "https://ai.gaobobo.cn/api/geo/verify-key" \
-H "Authorization: Bearer $(cat ~/.qclaw/geo-api-key)"
If the response code returns as anything other than 0, the process will halt automatically to protect credits.
Finalized packages and archives are stored dynamically using clean taxonomy paths containing brand names, products, and campaign cycles:
~/.qclaw/geo-exports/
└── <brand>_<product>_C<cycle>/
└── fangxie/
└── fangxie.zip
| Parameter | Type | Required | Description |
|---|---|---|---|
optimization_task_id |
String | Yes | The system task ID obtained automatically via the workflow coordinator |
cycle_number |
Integer | Yes | The current content generation loop index |
brand_name |
String | Yes | Target brand name utilized during drafting |
keyword |
String | Yes | Core keyword driving the copy adaptation |
reference_url |
String | Yes | Original content source URL |
reference_markdown |
String | Yes | Cleaned Markdown body extracted from source URL |
reference_outline |
Object | Yes | Extracted JSON structural layout with required section lists |
Loading
An automated workflow engine for QClaw that executes scheduled GEO optimization tasks, deep article imitations, and template exports natively in the background.

A comprehensive, non-destructive image editing toolkit powered by ImageMagick and packaged for automated AI development workflows.

A hybrid search tool querying local Openclaw documentation, skills, and wikis using BM25, Vector search, and LLM re-ranking.

An automated formatting checker and corrector designed to align Chinese academic papers with official university template requirements.

A step-by-step setup guide to seamlessly integrate Feishu (Lark) capabilities—including Docs, Calendars, Wiki, OKRs, and Messages—into your workspace using lark-cli.

A powerful single-command skill that generates a comprehensive whole-site AEO/GEO audit and an execution plan for your AI agent to optimize citation eligibility.








































