// A pattern for generating higher-quality output by iterating against explicit scoring criteria. Use for headlines, CTAs, landing page copy, social content, ad copy — anything where quality matters. Generate → Evaluate → Diagnose → Improve → Repeat.
Create and manage Privacy.com virtual cards for agent spending. Use when an agent needs to make a purchase, buy a domain, pay for a service, or needs a disposable card with a spending limit. Requires a Privacy.com account and API key.
Generate type scales, color palettes, spacing systems, WCAG contrast checks, and dark mode derivations with math. Use when setting up a design system, creating tokens, or building a Tailwind/CSS theme. Outputs CSS custom properties, Tailwind config, or JSON tokens.
Look up app details, ratings, reviews, and search the iOS App Store, Mac App Store, and Google Play. Use when the user asks about app ratings, wants to compare apps, or research a competitor's app store presence. No API key required.
Image prompt templates, model selection guidance, and anti-generic patterns for generating visual assets. Use when the user needs AI-generated images for landing pages, marketing, or products. Covers hero images, feature illustrations, OG cards, icons, and backgrounds.
Search and monitor Hacker News stories, comments, and users via the free Algolia API. Use when the user asks about HN discussions, wants to find posts about a topic, or monitor HN for mentions. No API key required.
Framer Motion (Motion) animation patterns for React — springs, staggers, layout animations, micro-interactions, scroll effects, and page transitions. Use when building or improving UI animations, adding polish, or making interfaces feel premium.
| name | recursive-improvement |
| description | A pattern for generating higher-quality output by iterating against explicit scoring criteria. Use for headlines, CTAs, landing page copy, social content, ad copy — anything where quality matters. Generate → Evaluate → Diagnose → Improve → Repeat. |
A pattern for generating higher-quality output by iterating against explicit scoring criteria.
generate → evaluate → diagnose → improve → repeat (until passing)
Never ship first-draft output for important content. Run the loop.
Create the initial output as you normally would.
Score the output against each criterion (1-10). Be brutally honest.
For any criterion scoring below threshold:
Rewrite addressing each diagnosed weakness. Don't patch — rebuild the weak sections.
Re-evaluate. Keep looping until all criteria pass threshold (usually 8/10 minimum).
After passing criteria, attack the output from a hostile perspective:
If it survives, ship it. If not, iterate.
| Criterion | What to evaluate |
|---|---|
| Hook strength | First line grabs attention? Pattern interrupt? |
| Curiosity gap | Creates urge to keep reading? |
| Clarity | One clear idea? No confusion? |
| Voice match | Sounds like the target voice/brand? |
| Engagement potential | People will reply/share/save? |
| Thumb-stop power | Scroller would pause? |
| Value density | Every line earns its place? |
| CTA clarity | Clear what reader should do next? |
Adversarial test: Would a distracted, skeptical user at 11pm engage with this?
| Criterion | What to evaluate |
|---|---|
| Headline clarity | Instantly clear what this business does? |
| Value prop strength | Why choose them over competitors? |
| Benefit focus | Features translated to customer benefits? |
| CTA effectiveness | Clear, compelling action? Low friction? |
| Trust signals | Credibility established? Social proof? |
| Readability | Scannable? Short paragraphs? Clear hierarchy? |
| Objection handling | Common concerns addressed? |
| Specificity | Concrete details vs vague claims? |
Adversarial test: Would someone searching on their phone take action within 30 seconds?
| Criterion | What to evaluate |
|---|---|
| Subject line | Would this get opened? Stands out in inbox? |
| Opening hook | First sentence earns the second? |
| Single focus | One clear ask per email? |
| Skimmability | Can get the gist in 5 seconds? |
| CTA prominence | Action is obvious and easy? |
| Voice consistency | Matches brand/sender personality? |
| Length appropriate | No fluff, nothing missing? |
| Mobile friendly | Works on small screens? |
Adversarial test: Would a busy person with 200 unread emails act on this?
| Criterion | What to evaluate |
|---|---|
| Thumb-stop power | Pattern interrupt in first 2 seconds? |
| Curiosity gap | Creates need to know more? |
| Emotional trigger | Hits a real pain point or desire? |
| Credibility | Believable? Not too good to be true? |
| CTA strength | Clear next step with low friction? |
| Persona match | Speaks directly to target audience? |
| Differentiation | Stands out from competitor ads? |
| Platform native | Fits the platform's style/format? |
Adversarial test: Would this stop YOUR scroll? Would you click?
Always use for:
Can skip for:
## Output v1
[Initial generation]
## Evaluation v1
- Hook strength: 6/10 — Opens weak, no pattern interrupt
- Clarity: 8/10 — Clear enough
- Voice match: 7/10 — Too formal
[... score all criteria]
## Diagnosis
1. Hook needs a surprising stat or contrarian take
2. Voice should be more casual, shorter sentences
3. [...]
## Output v2
[Revised version addressing weaknesses]
## Evaluation v2
[Re-score — continue until all pass]
The loop typically adds 2-3 iterations. Worth it for anything that matters.