THE COMPLETE INDEX
Follow a
better question.
Practical business help and deeper engineering, organized around the decision you need to make.
Run the business ↗
Useful AI automation for owners wearing many hats: choose the right job, protect customer trust, and measure the time and money that actually change.
02 / 6 NOTESAI discovery ↗
Make useful work easier for people and answer engines to find, understand, and cite.
03 / 1 NOTESAudience signals ↗
Questions, underserved needs, and testable opportunities at the edge of the work.
04 / 3 NOTESApplied craft ↗
Practical methods for publishing, knowledge environments, and editorial interfaces.
05 / 1 NOTESEvidence & measurement ↗
What changed, what can be measured, and what remains uncertain.
06 / 18 NOTESAgent systems ↗
Orchestration, MCP, memory, GraphRAG, and local inference. Decisions before novelty.

All field notes
An AI crawler fetched the page. What did that actually prove?
2 MIN ↗The client approved the image. Will its provenance survive delivery?
4 MIN ↗Keep the documents local. Make the retrieval earn the setup.
2 MIN ↗Your MCP server knows who called. Does it know what they may do?
3 MIN ↗Claude Enterprise starts at 20 seats. Check the work before the plan.
3 MIN ↗Your agents finished the job. What did the bad answer touch?
5 MIN ↗Give your team room to use AI without leaving the budget open.
3 MIN ↗Your agent keeps resending the past. Claude now has a shorter route.
3 MIN ↗Your PR bot may pause on 9 November. Check its trigger first.
3 MIN ↗Want AI to find your site, but not train on it? Keep those choices separate.
5 MIN ↗Gemini 4 Carbon is available. Is switching worth it for your team?
3 MIN ↗An AI reply can succeed and still contain the wrong customer's data.
3 MIN ↗Notion's credit-change date has passed. Check what your workspace actually uses.
3 MIN ↗Measure the handoff before you multiply it.
5 MIN ↗Your shop shows one price. Google found another.
5 MIN ↗A second agent should earn the handoff.
6 MIN ↗Three people sent feedback. Which version are you building?
6 MIN ↗You answered that question again. Give it a useful home.
6 MIN ↗The site is live. Can the client run it without you?
9 MIN ↗Before you add another agent, find the bottleneck.
6 MIN ↗The client said yes. Why can’t the work start?
6 MIN ↗A handoff is not done when the message is sent.
6 MIN ↗Your agent remembered. Did it change the next action?
4 MIN ↗You appeared in an AI answer. What did you actually measure?
10 MIN ↗Before you buy another Mac, measure the whole request.
6 MIN ↗Your next audience is describing the blockage, not your category.
11 MIN ↗The instructions still load. The advice may be wrong.
9 MIN ↗Before you automate, find the job holding the business back.
5 MIN ↗Give every enquiry a next step. Keep a human on the reply.
5 MIN ↗Your agent does not need every tool in the room.
4 MIN ↗Your agent remembers the answer. Who tells it the answer changed?
6 MIN ↗Did that automation actually give you time back?
6 MIN ↗A comparison page should help someone rule an option out.
4 MIN ↗Your best project is invisible if nobody can explain it.
9 MIN ↗Before you add a graph, find the missing answer.
3 MIN ↗SEO, AEO, GEO: what actually changes in the build?
10 MIN ↗FROM A GOOD READ TO A USEFUL NEXT STEP
Put one idea
to work.
Pick the problem on your desk. Try the method on the page, or take the whole kit.
The method is public. The kit needs one email address.