site-update
Skill16 starsChanged 44 days agoMITView on GitHub
Update sophonfinance.com (docs/index.html on GitHub Pages) with new sections or graphics. Use for any marketing-site change.
In this repository15 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- apSkill
- atlasSkill
- brainSkill
- brand-graphicsSkill
- build-loopSkill
- closeSkill
- linkedin-postSkill
- loopsSkill
- reconSkill
- shipSkill
- surplusSkill
- tax-1065Skill
- triangulateSkill
- validateSkill
The file
Files
sophonfinance.com Updates
The site is served by GitHub Pages from main: docs/index.html is the landing page, and generated engine datasheets live under docs/engines/. Changes go live only when merged to main.
Conventions
- Sections:
<section class="alt" id="...">on wash background;.sec-kickereyebrow,h2,.lead. Scroll-reveal via.revealclasses (IntersectionObserver adds.in). - Palette vars in
:root:--navy #0f2a5c, --blue #2563eb, --sky #00a8ff, --slate, --wash. - Trap:
.flow-captionis styled for DARK bands (light text). On light sections use inline dark styles instead. - Assets must live under
docs/assets/(Pages serves only /docs). Nav links are in the.nav-linksblock. - Engine datasheets are generated from
site-datasheets/specs/; update the source spec or shared partials and regenerate instead of hand-editingdocs/engines/*.html. - The "Self-Healing Operations" section (
#loops) embedsdocs/assets/self-healing-loop.svg.
Verify before pushing
Extract the section + <style> into a temp page, force-reveal (.reveal{opacity:1!important;transform:none!important}), render with headless chromium FROM the docs/ directory (so relative asset paths resolve), and Read the PNG. Check tag balance (sections/divs open == close).
Guardrail
The site promises "0 Numbers Finalized Without Approval" and "Humans stay the gate". Frame autonomy as governed autonomy (self-heal routine drift, escalate judgment, client sets the gate) unless the user explicitly re-positions — and if they do, the trust bar + principles sections must be rewritten too so the page doesn't contradict itself.
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.

