content-engine
CLAUDE.md9 starsChanged 6 days agoMITView on GitHub
This folder writes SEO articles for my site. Claude does the research, briefs, drafts and publishing. I review everything before it goes live. @site/site.md @site/voice.md
The file
We split this CLAUDE.md into the rules it gives an agent
3 rules. About 0 are common, 2 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 20
Every number, stat or claim in an article must come from: the data brief, my notes, or a page you fetched and will link to.
- Distinctive: line 26
Write like you're explaining it to a friend who asked.
- Distinctive: line 30
Use the PAA questions from the brief where they fit naturally.
≈ “Common” is an estimate: at least 1 in 5 of the 27,641 CLAUDE.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Content engine
This folder writes SEO articles for my site. Claude does the research, briefs, drafts and publishing. I review everything before it goes live.
@site/site.md @site/voice.md
The pipeline
/keywords <seed>pulls keyword data and turns it into a queue indata/queue.csv/brief <keyword>pulls the live Google results + the top pages and writesbriefs/<slug>.md- I answer the questions in
notes/<slug>.md(my experience, photos, opinions, numbers) /write <slug>writesdrafts/<slug>.md, then theeditoragent tears it apart and the fixes get applied/publish <slug>sends it to WordPress (as a draft unless I say otherwise)/autopilot <n>runs steps 2 to 5 for the next n keywords in the queue
Data rules
- Get SEO data only through
node scripts/dfs.mjs. It saves small files. Read those. Never readdata/raw/, it's huge and burns usage for nothing. - Every number, stat or claim in an article must come from: the data brief, my notes, or a page you fetched and will link to. If you don't have it, write
[NEED SOURCE: what]instead of guessing. - Anything that needs my first-hand experience and isn't in my notes gets
[ADD: what I should add]. Never invent experiences, tests, quotes, prices or "experts say".
Writing rules
- Answer the search in the first two or three sentences. No warm-up intro.
- Write like you're explaining it to a friend who asked. Plain words, short paragraphs, specific examples.
- Banned: em dashes, "delve", "in today's fast-paced world", "unlock", "game-changer", "navigate the landscape", "it's important to note", "whether you're a beginner or", "in conclusion", "elevate", "seamless".
- Length is whatever it takes to beat the top pages on usefulness. Never pad to hit a word count.
- 2 to 4 internal links to pages in
site/site.md, only where they actually help the reader. Link the external sources you used. - Use the PAA questions from the brief where they fit naturally. A short FAQ at the end only if there are questions left over.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

