{the brain} + {the skills}
Every output is the next input
{the system} is {the brain} plus {the skills}, run as one loop. Each skill reads the file the previous step wrote.
- skills in the loop
- 5
- steps, ask to review
- 7
- shared paths
- 9
01 · Three layers
Memory, steps, and the loop between them
Each layer works alone. Together they compound.
02 · The loop
Every outcome is the next input
Each skill reads a file, does one thing, and writes the next file. That file is the contract with the next step. Two steps are handwork, and the loop says so.
artifact
request
A request arrives. "We need a new dashboard."
artifact
docs/interview-guide.md
the request sharpened, plus the questions for the interviews
handwork
interviewing stakeholders and users
no skill. You talk to stakeholders and users
artifact
transcripts
in research/raw/
artifact
research/personas/
role description, workflows, job stories, pain points, needs, and more
skill
artifact
research/insights/
insights I01, I02 with evidence from the transcripts. Contradictions as C01
skill
artifact
docs/design-brief.md
the design assignment: scope, requirements, and which insights the design must serve
handwork
designing
no skill. You do this, in Figma
artifact
design
screens and a prototype, each screen naming the insight it serves
artifact
two reports
Standards: heuristics, WCAG. Evidence: which insight does this screen serve, and where is that written
- skill
- artifact, what the skill writes
- handwork, you do this yourself
Outside the loop
Outside the loop. Run at any point, from whatever the engagement folder holds. No loop step reads the file back.
- /euxlab-write-okrs
The engagement folder into two or three outcome OKRs, every key result with a baseline, a target, and a source.
03 · The state contract
9 paths, each with a writer and a reader
Every skill reads and writes the same folders. Change the contract and you change the loop.
| Path | Written by | Read by | Rule |
|---|---|---|---|
| docs/agents/engagement.md | /euxlab-setup | all | client, design system, consent posture, stakeholders |
| GLOSSARY.md | /euxlab-grill-the-ask, persona skill | all | a term earns an entry when it is fuzzy, contested, or overloaded |
| docs/interview-guide.md | /euxlab-grill-the-ask | persona skill, /euxlab-synthesize | the sharpened request, the graded assumption register, the questions |
| research/raw/ | interviewing | persona skill, /euxlab-synthesize | immutable, consent-checked, may contain PII |
| research/personas/ | persona skill | /euxlab-synthesize, /euxlab-design-brief | role, workflows, job stories, pain points, needs. No insights. Provisional until synthesis reconciles |
| research/insights/ | /euxlab-synthesize | /euxlab-design-brief, /euxlab-design-review | one file per insight, ID’d, graded, cited to raw |
| docs/design-brief.md | /euxlab-design-brief | designing, /euxlab-design-review | requirements, each serving named insight IDs |
| the design | designing | /euxlab-design-review | every screen names the insight ID it serves |
| docs/okrs.md | /euxlab-write-okrs | the team, at period end | outcome key results, each with a graded baseline, a target, and a source |
No brain yet?
Build the memory layer first
{the brain} is a free 7-day email course that builds the project memory. One email per day, one action per day.
{the brain}: free 7-day email course
Set up an Obsidian project brain that an LLM maintains for you. Seven days, one concrete action per day, 20 to 30 minutes each. Day 1 lands tomorrow morning.
No spam, no "value ladder". Seven emails, then you decide if you want more.
Hosted by Kit. Its script stores a visitor id in your browser.