U
⊨
22A8
models
▾
Score content against quality models
README scoring
8 traits learned from the best READMEs on GitHub. Hook speed, credibility, scannability.
Technical writing
Clarity, precision, structure. Score docs, tutorials, and blog posts across 6 traits.
Pun scoring
Semantic surprise, comedic density, absurd imagery — scored with a straight face.
Landing pages
Does your dev tool page earn trust? Technical depth, honest scope, zero-friction try.
plugins
▾
Use ⊨ in your editor
Claude Code plugin
Score and iteratively improve content until traits converge. Zero-config.
copy
$
claude mcp add \
--transport
http \ u22a8
https://u22a8.ai/mcp
mcp.json — works with any MCP client
copy
{
"mcpServers"
: {
"u22a8"
: {
"type"
:
"http"
,
"url"
:
"https://u22a8.ai/mcp"
} }}
docs
▾
Understanding scores
Models
What a resonance model is and why it matters.
Score card
Anatomy of a score — what every element means at a glance.
Traits
Independent dimensions of quality, each scored separately.
Tiers
Weak, Developing, Solid, Strong — what the labels mean.
Confidence
How reliable is this score? Three levels, and why Weak is high.
Integrate
REST API
Score content from your terminal. Pipe text, URLs, or files.
MCP
Connect any MCP client for scoring from your editor or agent.
GitHub Action
Score docs on every PR. Trait breakdown as a comment.
punchline
☰
U
⊨
22A8
×
Models
README scoring
8 traits from the best on GitHub
Technical writing
Clarity, precision, structure
Pun scoring
Scored with a straight face
Landing pages
Earn developer trust
Compare models
Side by side
Plugins
Claude Code plugin
Score & improve from your editor
MCP — paste into mcp.json:
copy
{
"mcpServers"
: {
"u22a8"
: {
"type"
:
"http"
,
"url"
:
"https://u22a8.ai/mcp"
} } }
Docs
Models
What a model is
Score card
Anatomy of a score
Traits
Dimensions of quality
Tiers
Weak to Strong
Confidence
How reliable is this score
Integrate
REST API
Score from your terminal
MCP
Connect any MCP client
GitHub Action
Score docs on every PR
Train
Coming soon
Overview
How a model learns your standard
Supervision
Four forms of input, one engine
Samples
Direct labeled examples
Briefs
Describe the standard
Discovery
Names the dimensions for you
Calibration
Breaks drawn from your data
Evolution
Re-train as your standard sharpens
punchline
You rate it. We rate it. Compare.
How good is this pun? Rate it.
50
Lock it in
You
vs
AI
.
.
.
Next pun
score your own →
0
Rounds
0
Avg gap
0
Bias
0
Streak
Sample puns are AI-generated.