inemaSEO 5 — With Agents: Claude Code and Codex Doing the Repetitive Work|INEMA.CLUB-PRO

inemaSEO Program · course 5 of 7 · 8 lessons

With Agents: Claude Code and Codex Doing the Repetitive Work

Audit, extract, apply in batches, check, and report: the repetitive part of SEO fits in an agent that runs through your subscription. In eight lessons, you build that line with the method's open scripts, an anti-fabrication check that discards what the AI made up, and two gates where only you decide.

Start with lesson 1 See the inemaSEO Area
A developer at a wooden desk jots a checklist in a notebook next to a laptop with the terminal open.
  • 8short lessons
  • 16exercises with "done when"
  • 2human gates
  • 0API calls without authorization

The lessons

In the order of the work: build the line, audit, extract, apply in batches, check, report, choose tools, and keep a record of what went wrong.

  1. Lesson 1 · 15 minThe machine with agents: who does what and where you decideAn AI agent is good at repetitive work: researching, drafting, checking, alerting, reporting. It is bad at deciding what is worth publishing. The inemaSEO method's machine keeps the two apart: the agents work in a line, and two gates stop everything until you approve.
  2. Lesson 2 · 18 minAuditor agent: the X-ray, claude-seo, and the list with evidenceThe first agent worth building is the one that only reads. It runs the X-ray, runs claude-seo, and returns a list of problems, each with the line that proves it. It fixes nothing: you decide what to change, and the 2026 yardstick filters out what's just a fad.
  3. Lesson 3 · 18 minExtraction through the subscription and the check that discardsListing pages, news article bodies, page summaries: the agent writes from a source, in batches. The risk is that it embellishes and makes things up. Here you call the model through the subscription, in read-only mode, read the first JSON in the reply, and run everything through a check that discards the made-up item.
  4. Lesson 4 · 17 minSafe batch operations: dry-run, marker, and dirty repositoryAdding a link back in forty repositories, swapping a footer on a hundred pages, fixing a canonical across the whole site: this is where the agent saves you hours, and this is where one mistake spreads. Three protections solve almost everything: show before you do, don't duplicate, and don't touch what someone else is working on.
  5. Lesson 5 · 17 minVerifiable goal in a loop, and the gate before the push"Make the page good" is not a goal: the agent decides on its own that it is done. A verifiable goal is a command with the expected output, plus a check that stops it from being met by a shortcut. The agent runs in a loop until it passes. After that, what no command can prove (first-hand data, voice, usefulness) is left for the human gate, before the push.
  6. Lesson 6 · 16 minAutomatic weekly report and the agent that reads it and proposesThe observer closes the machine: every week, a script gathers non-branded Search Console data, the indexing of the key pages, and the position of the target queries into one report. An agent reads it and proposes three actions, each with the line that justifies it. You approve what becomes a task.
  7. Lesson 7 · 16 minOpen vs. paid tools: what you can build for free and when to payA paid SEO tool sells data: search volume, competitor SERPs, backlinks, AI monitoring. For your own site, almost everything you need day to day comes from free sources: Search Console, Bing Webmaster, the server log, a local crawler, and the open Common Crawl graph. This lesson teaches you to tell the two apart and to decide when to pay.
  8. Lesson 8 · 20 minWhat went wrong with agents and the security rulesAgents fail in predictable ways: they claim without checking, add up wrong, break what was working, use the key they found. Each real failure in INEMA's SEO project became one line in the failure log and one rule in AGENTS.md. This lesson shows the failures, the smallest fix for each, and closes the course with your plan.

Who it's for

  1. 1People who finished courses 1 to 3 of the program and want to take the repetitive work off their hands.
  2. 2Technical people or devs who already use the terminal and want Claude Code or Codex to do SEO safely.
  3. 3People who have already tried leaving an AI "doing SEO on its own" and watched it make things up, duplicate, or touch things it shouldn't.

What you'll be able to do

  1. 1A line of agents with two human gates and an AGENTS.md with the project's rules.
  2. 2Audit and extraction through the subscription, with an anti-fabrication check that discards the made-up item.
  3. 3Batch operations with dry-run, marker, and skipping a dirty repository, and push only after the gate.
  4. 4A weekly report read by an agent that proposes, and you approve.

Where this course fits in the inemaSEO Program

  1. 11 Foundation, 2 Content, and 3 Measurement: the groundwork, pages with first-hand data, and the number that matters. This course automates what they teach you to do by hand.
  2. 24 In the AIs: official AI dashboards and the question battery.
  3. 35 With Agents (this course): agents doing the repetitive work, with gates. 6 Professional: the method as a service for clients. 7 Systems: the 5 open tools of inemaseo-sistemas, your own open Semrush.
  4. 4All of them follow the inemaSEO method, open (MIT), applied on the INEMA portal, news site, and events site.