Loslegen
Aether Engine: Find Demand and Publish SEO Pages

Aether Engine: Find Demand and Publish SEO Pages

Aether Engine helps founders and small marketing teams get found by people already looking. Your agent finds public posts that match what you sell, turns keyword data you import into a ranked page plan, writes long-tail SEO pages as static HTML with internal links and structured data, audits the files, and checks each draft for filler. Five standard-library Python scripts: one reads public sources, four are local only. Nothing is posted or deployed for you, and no ranking is promised.
#Marketing#Schreiben
Rating
Weitere Bewertungen erforderlich
Sold
0
How to use
Herunterladen

Aether Engine is for founders and small marketing teams who want to be found by people already looking. Your agent fetches public posts where people ask for what you sell and scores them against intent phrases you agree on, turns keyword data you import into a ranked page plan, writes long-tail SEO pages as static HTML, audits the files, and measures each draft for filler. The free Aether Listen skill covers only the first step. This product adds keyword planning, page publishing, the audit and the writing check.

What you get

  • scripts/listen.py (network only): fetches recent public posts from Hacker News, Reddit, StackExchange and RSS feeds you name. Reads no files and writes none.
  • scripts/signals.py (local only): scores posts against your intent phrases, groups the questions asked, and holds reply drafts until you approve or reject them. Reads aether.json, writes .aether/aether.db.
  • scripts/plan.py (local only): imports keyword JSON, CSV exports or plain lists, ranks and groups them into page topics, and prints a writing brief. Writes the same database.
  • scripts/site.py (local only): stores pages, writes them as HTML with Article and FAQ structured data, internal links, index.html, robots.txt and, when you set the site address, sitemap.xml, then audits them.
  • scripts/proof.py (local only): reads one draft and reports stock phrases, long sentences, repetition and a score. Writes nothing.
  • SKILL.md and two reference files: the agent's instructions, the aether.json format and the keyword import formats.

Requirements and install

Python 3.8 or newer (tested on 3.13). Standard library only: no pip installs, no API keys, no accounts, no git. Unzip the download: you get one folder named aether-engine. Put that folder in your agent client's skills folder. It is written for agent clients that read SKILL.md and can run a shell command, such as Claude Code or Codex CLI.

How to use

  • "I sell invoicing software for freelancers. Find people asking about unpaid invoices on Hacker News and show me the best matches."
  • "Add these keywords and show me the page plan: invoice template for freelancers, late payment reminder email, how to chase an unpaid invoice."
  • "Write the page for the top topic, check it, publish it and audit the site."

Sample run

From the publisher's test on 12 Nov 2026 (path shortened, output trimmed). keywords.json was a sample file in DataForSEO's format written for the test, not data from that service.

$ python <skill folder>/scripts/plan.py import keywords.json dataforseo
{
 "added": 4,
 "updated": 0,
 "total": 4
}
$ python <skill folder>/scripts/site.py audit
{
 "pages": 2,
 "errors": 0,
 "warnings": 0,
 "orphans": 0,
 ...
}

Output format

Every script prints JSON, and an error is one JSON line on standard error. The agent shows results in the chat as tables and text: top posts with link, score and how each was scored, the page plan with score, volume, difficulty and intent, and the audit counts. Pages are static HTML files in your project's site/ folder, and the chat summary names each one.

Network and data

Only listen.py uses the network. It sends read-only GET requests to hn.algolia.com, www.reddit.com, api.stackexchange.com and the feed URLs you add, and follows the redirects they return. A request carries your search words, subreddit, site or tag names and a fixed User-Agent. No login, key or cookie, and nothing from your files is sent. The posts it collects are public and are stored only in .aether/aether.db on your machine, as the poster wrote them: they include the poster's public name and anything else they published. Keyword numbers come from files you import: the scripts call no keyword or AI service. Your agent client sends the conversation, including command output, to the model provider you use.

What is automatic and what asks first

Automatic: fetching public posts, scoring, planning, and site.py publish, which writes files only into the local site folder. It replaces its own earlier output and stops with an error if a file it did not write is in the way. Nothing is deployed, uploaded, posted or sent, so hosting is your step. Replies are drafts: you approve each one and post it yourself. The agent asks before it writes aether.json, replaces or removes a page, publishes over files it did not write, or runs a paid keyword lookup. There is no undo command: keep each page's .json file to restore it.

Marks it leaves: every generated file carries the word aether-site (a generator tag in each page, a comment in robots.txt and the sitemap). A buy link gets utm_source=aether, utm_medium=seo and the page name as utm_campaign, unless you turn that off in aether.json.

Limits

  • Public posts only. The first ranking is phrase matching, so expect launch posts and job seekers until you add negative phrases. The intent model starts only after 20 decisions with at least 5 approvals and 5 rejections.
  • Hacker News is searched by your words. For Reddit, StackExchange and feeds the newest posts are read and your phrases pick the relevant ones. Reddit may rate-limit anonymous requests; a failed source is reported, not retried.
  • No keyword data is included. Search volume and difficulty are only what you import. Where one is missing the plan says so, and the score then rests on an assumption, not on measured demand.
  • Topics are grouped by shared words, not by search results.
  • The writing check is English only and measures patterns, not truth.
  • No ranking, traffic or sales is promised. Pages and replies are AI-assisted: label them where a platform or law requires it.

FAQ

Does it post replies or upload my site? No. No script can post, send or deploy.

Do I need a paid SEO tool? No. It imports Semrush, Ahrefs or Search Console CSV exports, DataForSEO-format JSON, or a plain list. Numbers it was not given stay unknown.

Will it write fake reviews or pose as a real person? No. The instructions tell the agent to refuse.

Licence

Proprietary, copyright (c) 2026 Abdullah Rathore: personal and non-transferable under the marketplace's end-user licence, and what you generate is yours (see LICENSE.txt).


Externe APIs

hn.algolia.com · www.reddit.com · api.stackexchange.com · RSS or Atom feed hosts you add