E-GEO skill
E-GEO scores a page on 10 GEO criteria, rewrites it and writes JSON-LD so AI search engines cite the site more often
- 218 ★ on the repository
- Last commit
- MIT
Extended access
Why this level
- Rewrites page text through a model on your API key and writes results into the `geo-output/` folder
- The scoring criteria include persuasion devices, so rewritten text needs an accuracy check
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/e-geoDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- Rewrites page text through a model on your API key and writes results into the `geo-output/` folder
- The scoring criteria include persuasion devices, so rewritten text needs an accuracy check
This is third-party code. Review the repository files before installing.
What it does
E-GEO is an open-source toolkit for optimizing content for AI search (GEO and AEO), meaning citation in answers from ChatGPT, Perplexity, Gemini and Claude. It consists of the `egeo` CLI and skills for Claude Code. It builds on the E-GEO preprint (arXiv:2511.20867), which supplies ten features: ranking emphasis, user intent, competitive edge, social proof, authority signals, scannability, citation density, structured data, factual language and freshness. The `/geo <url>` command runs a page through the full pipeline: analysis, scoring, rewriting, markup. The output is a `geo-output/` folder with `report.md`, `analysis.json`, the rewritten text in `optimized/`, JSON-LD in `schema/` and an implementation checklist. The other commands do parts of that path: `/geo:audit` only analyses, `/geo:optimize` handles a local file, `/geo:batch` walks a folder, `/geo:compete` compares with competitors, `/geo:report` builds a report and `/geo:loop` runs one iteration of a domain loop. In the CLI, `egeo optimize` does the same without an agent, and `egeo fix-gaps` takes an AI-visibility tracker export (or the output of `geo citations`) and rewrites the pages that lose a query. With `GEO_EVAL_MOCK=1` the commands run on a stub with no API key. The skills live in `.claude/skills/`: competitive-analysis, content-scoring, geo-loop, schema-generator and validation-doctor.
Who it is for. Site owners, SEO specialists and content teams who want ChatGPT, Perplexity and Gemini to name their pages in answers and are ready to edit text from a checklist.
Good fit when
- A pricing page is not cited anywhere in AI answers and you need a score on ten criteria with prioritised fixes
- A visibility tracker showed that a page loses on queries, and `egeo fix-gaps` rewrites the losers
- A folder of dozens of Markdown pages has to be optimized as a batch through `/geo:batch`
- A text needs JSON-LD markup: the schema-generator skill drafts it from the page
- You want to compare your page with competitors using `/geo:compete <query>`
Not a fit when
- You need a classic Google and Yandex search audit: the toolkit targets citation in AI answers
- Pages may not be rewritten by a model: the optimizer outputs finished text in place of yours
- You have no OpenAI-compatible API key: without one only the `GEO_EVAL_MOCK=1` trial run works
Example request
Run /geo on the pricing page, show the ten-criteria score and put the rewritten text and JSON-LD into geo-outputLimitations
- The effect on citations is not guaranteed. The author measures itself on 10 Perplexity queries and calls the result (5 of 10 in the 21 September 2026 snapshot) one snapshot, not a trend.
- The scoring criteria include persuasion devices (ranking words, urgency), so check rewritten text for accuracy before publishing.
- A real run needs an OpenAI-compatible API key; without one the stub trial mode works.
- Skills install through `npx skills add`, while the CLI installs separately with `pip install egeo`.
How to disable. Delete the skill folders from `.claude/skills/` (or with the command paired to `npx skills add`) if you installed through skills.sh, and remove the CLI with `pip uninstall egeo`. `geo-output/` folders in projects are deleted by hand.
FAQ
How do I install E-GEO in Claude Code?
Run `npx skills add https://github.com/mverab/eGEOagents` for the whole collection, or copy the `.claude/` folder into the project, then switch on `/output-style geo-optimizer` and run `/geo https://your-site`.
What are GEO and AEO?
GEO (Generative Engine Optimization) and AEO (Answer Engine Optimization) mean optimizing content for citation in AI search answers: ChatGPT, Perplexity, Google AI Overviews, Claude, Gemini. They are a set of techniques, not a new kind of link.
Can I try it without an API key?
Yes. `pip install egeo`, then `GEO_EVAL_MOCK=1 egeo optimize your-page.md` runs a trial on a stub and creates `geo-output/` with a score, a rewritten page, JSON-LD and a report.
How is E-GEO different from geo-optimizer?
geo-optimizer scores how ready a site is for citation in ChatGPT, Perplexity and AI search, as a CLI, library and MCP server. E-GEO goes past scoring: it rewrites the page, writes JSON-LD and closes gaps from a tracker export.
What does loop mode do?
One bounded pass over a workspace domain: `/geo:loop <domain>` in the agent or `egeo loop` in the CLI with run, collect and doctor subcommands. It is opt-in, and plain `/geo` and `egeo optimize` do not depend on it.
Discussion
No questions or reviews yet.
Related
SkillsA CLI, Python library and MCP server that scores how ready a site is to be cited by ChatGPT, Perplexity and AI Overviews
Three GEO skills: a one-off page audit from a URL, step-by-step optimization and a platform for long-term monitoring.
A Claude Code skill set for auditing sites for AI search: citability, AI crawler access, llms.txt, schema and reports
SEO and AI-search optimization skill: a 100-point site score, an action plan, ready fixes and 20 diagnostic scripts.
24 skills for GEO: page audits for AI answers, articles and comparisons, brand mention sampling in ChatGPT and DeepSeek, Word and PDF reports
A skill and CLI for GEO: check whether AI models can read your site, find who they recommend instead, draft fixes and re-measure.
Comparison with similar tools
Cards that share the main search query, and how they differ.
| Tool | Official | Repository stars | Main difference |
|---|---|---|---|
| E-GEO (this card) | no | 218 | |
| GEO Optimizer | no | 1,317 | a CLI, library and MCP server that scores a site's readiness for AI search citation |
| GEO-SEO Claude | no | 11,662 | a Claude Code skill set for auditing a site for AI search |
README in short
The README opens with a 30-second trial and a description of `egeo fix-gaps` for pairing with visibility trackers. It then explains what GEO is, how to install the set in Claude Code or through skills.sh, and lists the `/geo` commands, the `geo-output/` structure and the research behind it. The CLI reference, loop mode and the repository layout with skills follow.
README badge
Original SKILL.md
--- name: content-scoring description: Score content against the 10 GEO criteria with evidence and prioritized fixes. Use when users ask to score, rate, evaluate, or estimate ranking strength. --- # Content Scoring Skill When scoring content for GEO optimization: ## The 10 GEO Criteria Score each criterion 0-10: | # | Criterion | What to Look For | |---|-----------|------------------| | 1 | **Ranking Emphasis** | "best", "top", "#1", superlatives, leadership claims | | 2 | **User Intent** | Direct answers, addresses specific needs, solves problems | | 3 | **Competitive Diff** | Unique advantages, "unlike others", differentiators | | 4 | **Social Proof** | Stats, testimonials, reviews, customer counts, ratings | | 5 | **Narrative** | Engaging flow, persuasive language, compelling story | | 6 | **Authority** | Expert tone, credentials, specific knowledge, confidence | | 7 | **USPs** | Clear unique value, what makes it special | | 8 | **Urgency** | Time limits, scarcity, "now", limited availability | | 9 | **Scannable** | Headers, bullets, short paragraphs, clear structure | | 10 | **Factual** | Verifiable claims, specific numbers, accurate info | ## Scoring Guide - **0-2**: Missing or severely lacking - **3-4**: Present but weak - **5-6**: Adequate, room for improvement - **7-8**: Good, minor improvements possible - **9-10**: Excellent, near optimal ## Output Format ``` ┌─────────────────────────────────────────────────────────────┐ │ 📊 GEO CONTENT SCORE │ ├─────────────────────────────────────────────────────────────┤