Decide what to write next by checking Search Console queries and competitors with an MCP-driven SEO research agent
A newsletter writer built a coding-agent workflow that starts from his own Search Console queries, checks demand and competitors through a keyword tool, and labels each idea as write, update, expand, test or ignore.
Evidence: The author reports this. We have not checked it beyond reading the source.
The business problem
A publisher with a growing archive needs to know what to write or update next without duplicating pages that compete with each other.
What was tried
The agent pulls non-branded Search Console queries, especially those ranking around positions 5 to 20, and maps them to existing pages in a local content index. It checks five to ten candidate queries in a keyword tool via MCP for volume, difficulty and intent, identifies recurring competitor domains and runs a second pass on gaps. Three subagents validate the starting territory, propose competitor gaps, and assign page ownership while flagging cannibalisation. Each opportunity gets a decision and results are saved to a dated folder.
What was reported (positive)
On the author's site the run recommended updating a post with 906 search clicks and nine attributed paid subscribers, expanding an existing post, and ignoring tool comparisons, which averaged about 0.34 paid subscribers per 1,000 search clicks versus about 8.2 for how-to posts. A research run cost under $1 in credits.
Limitations
The post is paid and the setup steps, prompts and agent files are behind the paywall. Volume and difficulty figures are estimates, the conversion numbers come from small samples, and the path for new sites with little data is not tested. Results may not generalise beyond the author's niche, and some decisions still need human judgement.
What you need
Search Console access, a keyword research tool with MCP access (the hosted plan used was about $10 a month), and a coding agent such as Claude Code. Agent costs are not stated.
Sources
- The AI Maker (Substack) ↗ Firsthand write-up, published August 27, 2026
Source published: August 27, 2026. Last reviewed here: October 11, 2026. Spot a mistake? Tell us.
Tools in this workflow
- Claude Code: Runs the research agentTry Claude Code
- Google Search Console: Starting queries and performanceTry Google Search Console
Related workflows
Set up Hermes Agent on a cheap VPS with Telegram, morning cron briefings and a reusable content research skill
A 14-day test of Hermes Agent on a small VPS, reached through Telegram, with memory files, a daily and weekly briefing, and custom skills for content research.
Chain Hermes agents through Telegram into a research, writing, editing and publishing team
A business-automation blog shows how to let several Hermes agents message each other in Telegram so a research agent can hand work to a writing agent in a content pipeline.
Find content gaps with a Claude Code agent and then cross-check against your live archive
A creator ran a published content gap agent with Perplexity and Firecrawl, got a scored gap report, then discovered her local files missed 37 published posts and the real problem was distribution.
Give Claude Code a website team of connectors: Search Console, DataForSEO, Screaming Frog and PostHog
A founder connects five MCP servers to Claude Code, each acting as an SEO analyst, market researcher, technical auditor or product analyst, with the prompts he uses, including a weekly rage-click report.