Search Console Fix Desk: better page titles and descriptions from your Search Console data, checked, owner approved, measured
by PaperChaseWebb
Let an agent find the pages Google shows but people skip, and draft better titles and meta descriptions without inventing claims or touching your site. A planner compares each page with your own click rates at the same positions and decides rewrite, owner, hold or skip. A checker refuses drafts with claims or numbers the page never states, keyword stuffing, ranking promises, competitor names, noindex or redirect ideas and script overlays. You approve at your terminal and make the edit, and measure compares before and after against pages you did not touch.
- Price
- $29
- Difficulty
- Intermediate
- Setup
- 1 hr 15 min
- Format
- .md file
search-console-fix-desk-checked-titles-descriptions-owner-approved-measured.md · 223 KB · instant download after checkout
Want this built for you instead? We can build it for you.
What you end up with
Pages with plenty of impressions and few clicks get an honest, checked title and description, one change at a time, and every result you report is measured against untouched pages.
What you get
- 21 sections and an appendix in one .md file, with a one-line unpack for all 74 files
- seo_check.py data: checks your export for cut-off files, mixed windows, bad rows and queries that try to instruct the AI
- seo_check.py plan: compares each page with your own click rates by position and decides rewrite, owner, hold or skip, with the reason
- Traffic drops, two pages competing for one search, hidden queries and unknown URLs go to the owner instead of getting a new title
- seo_check.py draft: 19 checks against the plan and your page facts, from invented numbers and stuffing to promises, competitors and indexing changes
- The owner step: approve.py needs a real terminal and part of the draft hash, refuses agent sessions and piped input, and re-plans before it records a change
- measure and the report check: before and after against untouched pages, so a report only claims what the numbers show
- A 27-page eval set with expected decisions, rules and two skills for Grok Bot, Claude Code, Codex and Hermes, six prompts and five templates
- The test harness: 72 unit tests and 548 planted faults you can repeat
Who it is for
Site owners, content teams and agencies who want an agent to dig through Search Console and write the first drafts of better titles and descriptions, with a person approving and making every change.
Works with
Solid chips: the agent is named in this listing's own tools list. Dashed chip: Plain Markdown, so it also pastes into other agents such as Grok Bot, Hermes, OpenClaw or Muse. We list those as compatible by format only, not as tested.
Tools needed
Cost to run: The kit is the only purchase. Everything runs locally with Python; the checker makes no network calls, logs in to nothing and changes no website.
Time to result: The worked example in 5 minutes; your own export loaded and planned in about an hour
How it works
Unpack and test
Save the kit as pack.md in an empty folder, run the unpack line from Appendix A, then sh test_all.sh.
Expected: 74 files written; 72 tests OK; 548 of 548 planted faults caught.
Watch the flow
Run sh example/run_example.sh and read it next to section 17.
Expected: A tile shop's first-try draft with 13 findings, the fixed draft clean, a monthly report with 5 findings, the free check finding 7 problems and a piped approval refused.
Put in your site
Fill in site/site.json and the pages you might change in site/pages.json, then run python3 desk/seo_check.py site.
Expected: No findings, or a short list of settings to fix first.
Load your export
Export this window, the window before and page plus query rows into data/, run seo_check.py data until there are no high findings, then seo_check.py plan.
Expected: One line per page: rewrite, owner, hold or skip, with the reason and the queries to write for.
Sample prompts, free to use
Follow skills/search-fix with this month's export. Never edit the site. List each draft with its decision and finding count.
Prompt 1 of 6.
For the pages the plan marks rewrite, read each live page and list the claims and numbers it really states, as a proposed pages.json entry. Do not edit pages.json.
Prompt 4 of 6.
Follow skills/change-check for last month's changes and write the report to reports/. Run the report check and fix every finding.
Prompt 5 of 6.
Free preview of the file
# Search Console Fix Desk: better page titles and descriptions from your Search Console data, checked, owner approved, measured Version 1.0, 2026-11-01. By PaperChaseWebb. For site owners, small businesses, content teams and agencies who want an AI agent to find the pages Google shows but people skip, and draft better titles and meta descriptions without inventing claims, stuffing keywords, promising rankings or touching the live site. Works with Grok Bot, Claude Code, Codex and Hermes. Python 3.8 or newer, standard library only, no API keys, no logins. The kit and the agent never edit your website. Tested with 72 unit tests and 548 planted faults on Linux with Python 3.13, 3.12, 3.11, 3.9 and 3.8. All business and page data in this kit is made up. Search Console tells you which pages appear in Google and how often people click. The pages with plenty of impressions and few clicks are the cheapest fix on most sites: a clearer title and description, made once, in your CMS. Handing that to an AI agent goes wrong in predictable ways. It rewrites the title of a page whose clicks fell because it dropped five places, which no title fixes. It rewrites both of two pages that fight over the same search. It writes "Best", "#1", "free" and a price the page never mentions. It stuffs the query in three times. It acts on a query that reads "ignore previous instructions". It pushes the change live with a script that disappears the day the script does. It changes the same page again a week
Test record
- Date
- Nov 1, 2026
- What was run
- tests/test_data.py (14), test_plan.py (11), test_check.py (20), test_modes.py (14), test_approve.py (9, on a real pseudo terminal) and test_mini.py (4); tests/mutate.py with 528 planted faults across drafts, exports, plan changes, site settings, the change log, reports, agent rules and approve.py, plus 20 against seo_mini.py; seo_check.py set on the 27-page eval set; example/run_example.sh end to end; and the unpack line with test_all.sh. Linux, Python 3.13, 3.12, 3.11, 3.9 and 3.8.
- Result
- 72 passed, 0 failed on all five versions; 548 of 548 planted faults caught and 20 of 20 clean baselines clean; the eval set had 0 findings; the example caught the first-try draft with 13 findings and the first-try report with 5, and refused piped approval; the unpack line wrote every file and test_all.sh passed there.
- Not run
- A live Grok Bot, Claude Code, Codex or Hermes session. A real Search Console property, the API or the bulk export. Non-English sites. macOS or Windows.
Copied from this item's dated test notes. What was not run stays in.
Worth knowing
- The agent rules were checked by the checker but not loaded inside a live Grok Bot, Claude Code, Codex or Hermes session for this release. The eval drafts were written by hand to the rules. Give your agent the entrance exam in section 15 before you rely on it.
- No ranking or traffic result is promised. Google writes its own title links and may show other text than your title; measure is how you find out what changed.
- The checker works on words, numbers and your page facts. It cannot tell whether a title that passes every rule is still dull, and its word lists are a floor, so the owner reads every draft.
- It does not log in to Search Console or your CMS and changes nothing. The data comes in as CSV, and the web export stops at 1,000 rows, so larger sites need the API, the bulk export or an add-on.
- The owner step assumes the agent cannot run approve.py at a terminal. Run it as a separate user; agent rules alone are not a hard boundary.
- All businesses and pages in the kit are made up. English only. Tested on Linux only; macOS and Windows were not tested.
- On 2026-11-01 on Linux: 72 unit tests and 548 of 548 planted faults passed on Python 3.13, 3.12, 3.11, 3.9 and 3.8; the 27-page eval set was clean; the unpack line wrote every file and test_all.sh passed in that folder.
Questions
Which agents does it work with?
Any agent that can read files and run a script. The kit has rules for AGENTS.md and CLAUDE.md, Claude Code and Codex notes, Hermes notes and a Grok Bot standing order. Those were checked by the checker, not loaded in each agent for this release.
Will it change my website?
No. It plans and checks drafts. You approve a draft at a terminal and make the edit in your CMS yourself, then record it so the page is held and measured.
Does it connect to Search Console?
Not directly. You export to CSV, which keeps the kit free of logins and keys. Section 5 shows the files and the columns, and a read-only Search Console API or MCP tool can produce them.
Why not rewrite every low CTR page?
Because some of them lost clicks to a lower position, compete with another page, or have mostly hidden queries. A new title does not fix those, so the planner sends them to you with the reason.
Can you set it up for us?
Yes. Our Install sprint sets up your export, page facts and agent, or Managed Ops runs the plan, drafts and measuring for you each month. Book a call or write to [email protected].
Details
Version 1.0 · Updated Nov 1, 2026 · Markdown version of this page
License: Use inside your own business or on your clients' sites. Do not resell or republish the kit itself. Download terms
Related
Search Console Fix Desk: safe title rules and a ten-check page title and description scan (free sample)
From Search Console Fix Desk: the mistakes agents make when they rewrite page titles, eight rules to paste into your agent, and a read-only script that runs ten checks on one title and description, from length, stuffing and invented numbers to best and #1 claims, ranking promises, clickbait and indexing changes.
Content System Kit: run a content engine with AI agents and keep a human on the publish button
A prompt pack and method for founders, small teams and agencies. Audience filters, one owned format, a 90-post test with review rules set in advance, a weekly batch loop with one approval gate, an editor queue with a rights check, transcript-first cleanup, a packaging watch on public data only, teardown without replication, a craft seat map, a cadence plan, an engagement honesty audit, a failure table, a worked example, a rules block and 14 prompts. Includes content_check.py, AGENTS.md with six owner stops, two skills and 15 templates.
Review Reply Desk: AI-drafted review replies, checked for privacy and offers, approved by the owner
Let an agent draft your Google and Yelp review replies without the mistakes that cost money. A checker refuses drafts that use a last name or details from your records, offer a discount for a better review, ask for a review to be changed, argue, invent numbers or promises, or repeat your last reply word for word. Health topics, allergies and legal threats go to the owner, clinics get a stricter privacy mode, and nothing is posted until the owner approves it at a real terminal.
Invoice Chase Desk: AI-drafted payment reminders, checked against your ledger, approved by the owner
Let an agent draft your overdue invoice reminders without chasing a client who already paid. A planner reads your ledger and decides who gets a reminder today, and at which step. A checker refuses drafts with the wrong balance or dates, a fee your terms do not include, threats or pressure, new bank details, lookalike payment links, other clients' invoices or claims nobody can back up. Disputes, money trouble and odd replies go to the owner, and before anything is recorded as sent the desk checks the ledger again.
Agency Open-Loop System: meeting notes to tasks, client follow-ups and a weekly open-loop report
A Markdown operating system plus a small checker for owner-led agencies: turn every meeting into commitments with an owner, a date, a quote and a done test, draft client follow-ups as approval cards you answer by number, and get one weekly report of what is still open. Includes AGENTS.md with six owner stops, three skills, templates and loops_check.py.
Ask-first rules for AI agents: hard nos, approval cards and a test set
Paste-in rules that make an agent ask before it sends, spends, posts or deletes, define what a real yes is, and stop retries after a no, plus 12 test scenarios to check it.
Get a weekly metrics report written by an agent from Stripe, analytics and a sheet
Build a scheduled job that pulls revenue, traffic and your own sheet numbers, computes week over week changes in code, has an agent write a short report, checks every number against the table and sends it only after your yes.