Mnemex
index.html•58.1 kB
<!doctype html>
<html lang="en" class="no-js">
<head>
<meta charset="utf-8">
<meta name="viewport" content="width=device-width,initial-scale=1">
<meta name="description" content="Memory persistence for AI assistants with temporal decay">
<meta name="author" content="prefrontal-systems">
<link rel="canonical" href="https://cortexgraph.dev/prompt_optimization_flow/">
<link rel="icon" href="../assets/images/favicon.png">
<meta name="generator" content="mkdocs-1.6.1, mkdocs-material-9.7.0">
<title>Prompt Optimization Flow - CortexGraph Documentation</title>
<link rel="stylesheet" href="../assets/stylesheets/main.618322db.min.css">
<link rel="stylesheet" href="../assets/stylesheets/palette.ab4e12ef.min.css">
<link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
<link rel="stylesheet" href="https://fonts.googleapis.com/css?family=Roboto:300,300i,400,400i,700,700i%7CRoboto+Mono:400,400i,700,700i&display=fallback">
<style>:root{--md-text-font:"Roboto";--md-code-font:"Roboto Mono"}</style>
<script>__md_scope=new URL("..",location),__md_hash=e=>[...e].reduce(((e,_)=>(e<<5)-e+_.charCodeAt(0)),0),__md_get=(e,_=localStorage,t=__md_scope)=>JSON.parse(_.getItem(t.pathname+"."+e)),__md_set=(e,_,t=localStorage,a=__md_scope)=>{try{t.setItem(a.pathname+"."+e,JSON.stringify(_))}catch(e){}}</script>
</head>
<body dir="ltr" data-md-color-scheme="default" data-md-color-primary="indigo" data-md-color-accent="indigo">
<input class="md-toggle" data-md-toggle="drawer" type="checkbox" id="__drawer" autocomplete="off">
<input class="md-toggle" data-md-toggle="search" type="checkbox" id="__search" autocomplete="off">
<label class="md-overlay" for="__drawer"></label>
<div data-md-component="skip">
<a href="#prompt-optimization-flow" class="md-skip">
Skip to content
</a>
</div>
<div data-md-component="announce">
</div>
<header class="md-header" data-md-component="header">
<nav class="md-header__inner md-grid" aria-label="Header">
<a href=".." title="CortexGraph Documentation" class="md-header__button md-logo" aria-label="CortexGraph Documentation" data-md-component="logo">
<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M12 8a3 3 0 0 0 3-3 3 3 0 0 0-3-3 3 3 0 0 0-3 3 3 3 0 0 0 3 3m0 3.54C9.64 9.35 6.5 8 3 8v11c3.5 0 6.64 1.35 9 3.54 2.36-2.19 5.5-3.54 9-3.54V8c-3.5 0-6.64 1.35-9 3.54"/></svg>
</a>
<label class="md-header__button md-icon" for="__drawer">
<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M3 6h18v2H3zm0 5h18v2H3zm0 5h18v2H3z"/></svg>
</label>
<div class="md-header__title" data-md-component="header-title">
<div class="md-header__ellipsis">
<div class="md-header__topic">
<span class="md-ellipsis">
CortexGraph Documentation
</span>
</div>
<div class="md-header__topic" data-md-component="header-topic">
<span class="md-ellipsis">
Prompt Optimization Flow
</span>
</div>
</div>
</div>
<form class="md-header__option" data-md-component="palette">
<input class="md-option" data-md-color-media="(prefers-color-scheme: light)" data-md-color-scheme="default" data-md-color-primary="indigo" data-md-color-accent="indigo" aria-label="Switch to dark mode" type="radio" name="__palette" id="__palette_0">
<label class="md-header__button md-icon" title="Switch to dark mode" for="__palette_1" hidden>
<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M12 8a4 4 0 0 0-4 4 4 4 0 0 0 4 4 4 4 0 0 0 4-4 4 4 0 0 0-4-4m0 10a6 6 0 0 1-6-6 6 6 0 0 1 6-6 6 6 0 0 1 6 6 6 6 0 0 1-6 6m8-9.31V4h-4.69L12 .69 8.69 4H4v4.69L.69 12 4 15.31V20h4.69L12 23.31 15.31 20H20v-4.69L23.31 12z"/></svg>
</label>
<input class="md-option" data-md-color-media="(prefers-color-scheme: dark)" data-md-color-scheme="slate" data-md-color-primary="indigo" data-md-color-accent="indigo" aria-label="Switch to light mode" type="radio" name="__palette" id="__palette_1">
<label class="md-header__button md-icon" title="Switch to light mode" for="__palette_0" hidden>
<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M12 18c-.89 0-1.74-.2-2.5-.55C11.56 16.5 13 14.42 13 12s-1.44-4.5-3.5-5.45C10.26 6.2 11.11 6 12 6a6 6 0 0 1 6 6 6 6 0 0 1-6 6m8-9.31V4h-4.69L12 .69 8.69 4H4v4.69L.69 12 4 15.31V20h4.69L12 23.31 15.31 20H20v-4.69L23.31 12z"/></svg>
</label>
</form>
<script>var palette=__md_get("__palette");if(palette&&palette.color){if("(prefers-color-scheme)"===palette.color.media){var media=matchMedia("(prefers-color-scheme: light)"),input=document.querySelector(media.matches?"[data-md-color-media='(prefers-color-scheme: light)']":"[data-md-color-media='(prefers-color-scheme: dark)']");palette.color.media=input.getAttribute("data-md-color-media"),palette.color.scheme=input.getAttribute("data-md-color-scheme"),palette.color.primary=input.getAttribute("data-md-color-primary"),palette.color.accent=input.getAttribute("data-md-color-accent")}for(var[key,value]of Object.entries(palette.color))document.body.setAttribute("data-md-color-"+key,value)}</script>
<label class="md-header__button md-icon" for="__search">
<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M9.5 3A6.5 6.5 0 0 1 16 9.5c0 1.61-.59 3.09-1.56 4.23l.27.27h.79l5 5-1.5 1.5-5-5v-.79l-.27-.27A6.52 6.52 0 0 1 9.5 16 6.5 6.5 0 0 1 3 9.5 6.5 6.5 0 0 1 9.5 3m0 2C7 5 5 7 5 9.5S7 14 9.5 14 14 12 14 9.5 12 5 9.5 5"/></svg>
</label>
<div class="md-search" data-md-component="search" role="dialog">
<label class="md-search__overlay" for="__search"></label>
<div class="md-search__inner" role="search">
<form class="md-search__form" name="search">
<input type="text" class="md-search__input" name="query" aria-label="Search" placeholder="Search" autocapitalize="off" autocorrect="off" autocomplete="off" spellcheck="false" data-md-component="search-query" required>
<label class="md-search__icon md-icon" for="__search">
<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M9.5 3A6.5 6.5 0 0 1 16 9.5c0 1.61-.59 3.09-1.56 4.23l.27.27h.79l5 5-1.5 1.5-5-5v-.79l-.27-.27A6.52 6.52 0 0 1 9.5 16 6.5 6.5 0 0 1 3 9.5 6.5 6.5 0 0 1 9.5 3m0 2C7 5 5 7 5 9.5S7 14 9.5 14 14 12 14 9.5 12 5 9.5 5"/></svg>
<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M20 11v2H8l5.5 5.5-1.42 1.42L4.16 12l7.92-7.92L13.5 5.5 8 11z"/></svg>
</label>
<nav class="md-search__options" aria-label="Search">
<button type="reset" class="md-search__icon md-icon" title="Clear" aria-label="Clear" tabindex="-1">
<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M19 6.41 17.59 5 12 10.59 6.41 5 5 6.41 10.59 12 5 17.59 6.41 19 12 13.41 17.59 19 19 17.59 13.41 12z"/></svg>
</button>
</nav>
<div class="md-search__suggest" data-md-component="search-suggest"></div>
</form>
<div class="md-search__output">
<div class="md-search__scrollwrap" tabindex="0" data-md-scrollfix>
<div class="md-search-result" data-md-component="search-result">
<div class="md-search-result__meta">
Initializing search
</div>
<ol class="md-search-result__list" role="presentation"></ol>
</div>
</div>
</div>
</div>
</div>
<div class="md-header__source">
<a href="https://github.com/prefrontal-systems/cortexgraph" title="Go to repository" class="md-source" data-md-component="source">
<div class="md-source__icon md-icon">
<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 512 512"><!--! Font Awesome Free 7.1.0 by @fontawesome - https://fontawesome.com License - https://fontawesome.com/license/free (Icons: CC BY 4.0, Fonts: SIL OFL 1.1, Code: MIT License) Copyright 2025 Fonticons, Inc.--><path d="M173.9 397.4c0 2-2.3 3.6-5.2 3.6-3.3.3-5.6-1.3-5.6-3.6 0-2 2.3-3.6 5.2-3.6 3-.3 5.6 1.3 5.6 3.6m-31.1-4.5c-.7 2 1.3 4.3 4.3 4.9 2.6 1 5.6 0 6.2-2s-1.3-4.3-4.3-5.2c-2.6-.7-5.5.3-6.2 2.3m44.2-1.7c-2.9.7-4.9 2.6-4.6 4.9.3 2 2.9 3.3 5.9 2.6 2.9-.7 4.9-2.6 4.6-4.6-.3-1.9-3-3.2-5.9-2.9M252.8 8C114.1 8 8 113.3 8 252c0 110.9 69.8 205.8 169.5 239.2 12.8 2.3 17.3-5.6 17.3-12.1 0-6.2-.3-40.4-.3-61.4 0 0-70 15-84.7-29.8 0 0-11.4-29.1-27.8-36.6 0 0-22.9-15.7 1.6-15.4 0 0 24.9 2 38.6 25.8 21.9 38.6 58.6 27.5 72.9 20.9 2.3-16 8.8-27.1 16-33.7-55.9-6.2-112.3-14.3-112.3-110.5 0-27.5 7.6-41.3 23.6-58.9-2.6-6.5-11.1-33.3 2.6-67.9 20.9-6.5 69 27 69 27 20-5.6 41.5-8.5 62.8-8.5s42.8 2.9 62.8 8.5c0 0 48.1-33.6 69-27 13.7 34.7 5.2 61.4 2.6 67.9 16 17.7 25.8 31.5 25.8 58.9 0 96.5-58.9 104.2-114.8 110.5 9.2 7.9 17 22.9 17 46.4 0 33.7-.3 75.4-.3 83.6 0 6.5 4.6 14.4 17.3 12.1C436.2 457.8 504 362.9 504 252 504 113.3 391.5 8 252.8 8M105.2 352.9c-1.3 1-1 3.3.7 5.2 1.6 1.6 3.9 2.3 5.2 1 1.3-1 1-3.3-.7-5.2-1.6-1.6-3.9-2.3-5.2-1m-10.8-8.1c-.7 1.3.3 2.9 2.3 3.9 1.6 1 3.6.7 4.3-.7.7-1.3-.3-2.9-2.3-3.9-2-.6-3.6-.3-4.3.7m32.4 35.6c-1.6 1.3-1 4.3 1.3 6.2 2.3 2.3 5.2 2.6 6.5 1 1.3-1.3.7-4.3-1.3-6.2-2.2-2.3-5.2-2.6-6.5-1m-11.4-14.7c-1.6 1-1.6 3.6 0 5.9s4.3 3.3 5.6 2.3c1.6-1.3 1.6-3.9 0-6.2-1.4-2.3-4-3.3-5.6-2"/></svg>
</div>
<div class="md-source__repository">
prefrontal-systems/cortexgraph
</div>
</a>
</div>
</nav>
</header>
<div class="md-container" data-md-component="container">
<nav class="md-tabs" aria-label="Tabs" data-md-component="tabs">
<div class="md-grid">
<ul class="md-tabs__list">
<li class="md-tabs__item">
<a href=".." class="md-tabs__link">
Home
</a>
</li>
<li class="md-tabs__item">
<a href="../installation/" class="md-tabs__link">
Getting Started
</a>
</li>
<li class="md-tabs__item">
<a href="../architecture/" class="md-tabs__link">
Documentation
</a>
</li>
<li class="md-tabs__item">
<a href="../deployment/" class="md-tabs__link">
Deployment
</a>
</li>
<li class="md-tabs__item">
<a href="../CONTRIBUTING/" class="md-tabs__link">
Development
</a>
</li>
<li class="md-tabs__item">
<a href="../features/auto-recall-conversation/" class="md-tabs__link">
Features
</a>
</li>
<li class="md-tabs__item">
<a href="../LICENSE/" class="md-tabs__link">
About
</a>
</li>
</ul>
</div>
</nav>
<main class="md-main" data-md-component="main">
<div class="md-main__inner md-grid">
<div class="md-sidebar md-sidebar--primary" data-md-component="sidebar" data-md-type="navigation" >
<div class="md-sidebar__scrollwrap">
<div class="md-sidebar__inner">
<nav class="md-nav md-nav--primary md-nav--lifted" aria-label="Navigation" data-md-level="0">
<label class="md-nav__title" for="__drawer">
<a href=".." title="CortexGraph Documentation" class="md-nav__button md-logo" aria-label="CortexGraph Documentation" data-md-component="logo">
<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M12 8a3 3 0 0 0 3-3 3 3 0 0 0-3-3 3 3 0 0 0-3 3 3 3 0 0 0 3 3m0 3.54C9.64 9.35 6.5 8 3 8v11c3.5 0 6.64 1.35 9 3.54 2.36-2.19 5.5-3.54 9-3.54V8c-3.5 0-6.64 1.35-9 3.54"/></svg>
</a>
CortexGraph Documentation
</label>
<div class="md-nav__source">
<a href="https://github.com/prefrontal-systems/cortexgraph" title="Go to repository" class="md-source" data-md-component="source">
<div class="md-source__icon md-icon">
<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 512 512"><!--! Font Awesome Free 7.1.0 by @fontawesome - https://fontawesome.com License - https://fontawesome.com/license/free (Icons: CC BY 4.0, Fonts: SIL OFL 1.1, Code: MIT License) Copyright 2025 Fonticons, Inc.--><path d="M173.9 397.4c0 2-2.3 3.6-5.2 3.6-3.3.3-5.6-1.3-5.6-3.6 0-2 2.3-3.6 5.2-3.6 3-.3 5.6 1.3 5.6 3.6m-31.1-4.5c-.7 2 1.3 4.3 4.3 4.9 2.6 1 5.6 0 6.2-2s-1.3-4.3-4.3-5.2c-2.6-.7-5.5.3-6.2 2.3m44.2-1.7c-2.9.7-4.9 2.6-4.6 4.9.3 2 2.9 3.3 5.9 2.6 2.9-.7 4.9-2.6 4.6-4.6-.3-1.9-3-3.2-5.9-2.9M252.8 8C114.1 8 8 113.3 8 252c0 110.9 69.8 205.8 169.5 239.2 12.8 2.3 17.3-5.6 17.3-12.1 0-6.2-.3-40.4-.3-61.4 0 0-70 15-84.7-29.8 0 0-11.4-29.1-27.8-36.6 0 0-22.9-15.7 1.6-15.4 0 0 24.9 2 38.6 25.8 21.9 38.6 58.6 27.5 72.9 20.9 2.3-16 8.8-27.1 16-33.7-55.9-6.2-112.3-14.3-112.3-110.5 0-27.5 7.6-41.3 23.6-58.9-2.6-6.5-11.1-33.3 2.6-67.9 20.9-6.5 69 27 69 27 20-5.6 41.5-8.5 62.8-8.5s42.8 2.9 62.8 8.5c0 0 48.1-33.6 69-27 13.7 34.7 5.2 61.4 2.6 67.9 16 17.7 25.8 31.5 25.8 58.9 0 96.5-58.9 104.2-114.8 110.5 9.2 7.9 17 22.9 17 46.4 0 33.7-.3 75.4-.3 83.6 0 6.5 4.6 14.4 17.3 12.1C436.2 457.8 504 362.9 504 252 504 113.3 391.5 8 252.8 8M105.2 352.9c-1.3 1-1 3.3.7 5.2 1.6 1.6 3.9 2.3 5.2 1 1.3-1 1-3.3-.7-5.2-1.6-1.6-3.9-2.3-5.2-1m-10.8-8.1c-.7 1.3.3 2.9 2.3 3.9 1.6 1 3.6.7 4.3-.7.7-1.3-.3-2.9-2.3-3.9-2-.6-3.6-.3-4.3.7m32.4 35.6c-1.6 1.3-1 4.3 1.3 6.2 2.3 2.3 5.2 2.6 6.5 1 1.3-1.3.7-4.3-1.3-6.2-2.2-2.3-5.2-2.6-6.5-1m-11.4-14.7c-1.6 1-1.6 3.6 0 5.9s4.3 3.3 5.6 2.3c1.6-1.3 1.6-3.9 0-6.2-1.4-2.3-4-3.3-5.6-2"/></svg>
</div>
<div class="md-source__repository">
prefrontal-systems/cortexgraph
</div>
</a>
</div>
<ul class="md-nav__list" data-md-scrollfix>
<li class="md-nav__item">
<a href=".." class="md-nav__link">
<span class="md-ellipsis">
Home
</span>
</a>
</li>
<li class="md-nav__item md-nav__item--nested">
<input class="md-nav__toggle md-toggle md-toggle--indeterminate" type="checkbox" id="__nav_2" >
<label class="md-nav__link" for="__nav_2" id="__nav_2_label" tabindex="0">
<span class="md-ellipsis">
Getting Started
</span>
<span class="md-nav__icon md-icon"></span>
</label>
<nav class="md-nav" data-md-level="1" aria-labelledby="__nav_2_label" aria-expanded="false">
<label class="md-nav__title" for="__nav_2">
<span class="md-nav__icon md-icon"></span>
Getting Started
</label>
<ul class="md-nav__list" data-md-scrollfix>
<li class="md-nav__item">
<a href="../installation/" class="md-nav__link">
<span class="md-ellipsis">
Installation
</span>
</a>
</li>
<li class="md-nav__item">
<a href="../quickstart/" class="md-nav__link">
<span class="md-ellipsis">
Quick Start
</span>
</a>
</li>
<li class="md-nav__item">
<a href="../configuration/" class="md-nav__link">
<span class="md-ellipsis">
Configuration
</span>
</a>
</li>
</ul>
</nav>
</li>
<li class="md-nav__item md-nav__item--nested">
<input class="md-nav__toggle md-toggle md-toggle--indeterminate" type="checkbox" id="__nav_3" >
<label class="md-nav__link" for="__nav_3" id="__nav_3_label" tabindex="0">
<span class="md-ellipsis">
Documentation
</span>
<span class="md-nav__icon md-icon"></span>
</label>
<nav class="md-nav" data-md-level="1" aria-labelledby="__nav_3_label" aria-expanded="false">
<label class="md-nav__title" for="__nav_3">
<span class="md-nav__icon md-icon"></span>
Documentation
</label>
<ul class="md-nav__list" data-md-scrollfix>
<li class="md-nav__item">
<a href="../architecture/" class="md-nav__link">
<span class="md-ellipsis">
Architecture
</span>
</a>
</li>
<li class="md-nav__item">
<a href="../api/" class="md-nav__link">
<span class="md-ellipsis">
API Reference
</span>
</a>
</li>
<li class="md-nav__item">
<a href="../graph_features/" class="md-nav__link">
<span class="md-ellipsis">
Knowledge Graph
</span>
</a>
</li>
<li class="md-nav__item">
<a href="../scoring_algorithm/" class="md-nav__link">
<span class="md-ellipsis">
Scoring Algorithm
</span>
</a>
</li>
</ul>
</nav>
</li>
<li class="md-nav__item md-nav__item--nested">
<input class="md-nav__toggle md-toggle md-toggle--indeterminate" type="checkbox" id="__nav_4" >
<label class="md-nav__link" for="__nav_4" id="__nav_4_label" tabindex="0">
<span class="md-ellipsis">
Deployment
</span>
<span class="md-nav__icon md-icon"></span>
</label>
<nav class="md-nav" data-md-level="1" aria-labelledby="__nav_4_label" aria-expanded="false">
<label class="md-nav__title" for="__nav_4">
<span class="md-nav__icon md-icon"></span>
Deployment
</label>
<ul class="md-nav__list" data-md-scrollfix>
<li class="md-nav__item">
<a href="../deployment/" class="md-nav__link">
<span class="md-ellipsis">
Deployment Guide
</span>
</a>
</li>
<li class="md-nav__item">
<a href="../security/" class="md-nav__link">
<span class="md-ellipsis">
Security
</span>
</a>
</li>
</ul>
</nav>
</li>
<li class="md-nav__item md-nav__item--nested">
<input class="md-nav__toggle md-toggle md-toggle--indeterminate" type="checkbox" id="__nav_5" >
<label class="md-nav__link" for="__nav_5" id="__nav_5_label" tabindex="0">
<span class="md-ellipsis">
Development
</span>
<span class="md-nav__icon md-icon"></span>
</label>
<nav class="md-nav" data-md-level="1" aria-labelledby="__nav_5_label" aria-expanded="false">
<label class="md-nav__title" for="__nav_5">
<span class="md-nav__icon md-icon"></span>
Development
</label>
<ul class="md-nav__list" data-md-scrollfix>
<li class="md-nav__item">
<a href="../CONTRIBUTING/" class="md-nav__link">
<span class="md-ellipsis">
Contributing
</span>
</a>
</li>
<li class="md-nav__item">
<a href="../ROADMAP/" class="md-nav__link">
<span class="md-ellipsis">
Roadmap
</span>
</a>
</li>
<li class="md-nav__item">
<a href="../future_roadmap/" class="md-nav__link">
<span class="md-ellipsis">
Future Plans
</span>
</a>
</li>
</ul>
</nav>
</li>
<li class="md-nav__item md-nav__item--nested">
<input class="md-nav__toggle md-toggle md-toggle--indeterminate" type="checkbox" id="__nav_6" >
<label class="md-nav__link" for="__nav_6" id="__nav_6_label" tabindex="0">
<span class="md-ellipsis">
Features
</span>
<span class="md-nav__icon md-icon"></span>
</label>
<nav class="md-nav" data-md-level="1" aria-labelledby="__nav_6_label" aria-expanded="false">
<label class="md-nav__title" for="__nav_6">
<span class="md-nav__icon md-icon"></span>
Features
</label>
<ul class="md-nav__list" data-md-scrollfix>
<li class="md-nav__item">
<a href="../features/auto-recall-conversation/" class="md-nav__link">
<span class="md-ellipsis">
Auto-Recall
</span>
</a>
</li>
</ul>
</nav>
</li>
<li class="md-nav__item md-nav__item--nested">
<input class="md-nav__toggle md-toggle md-toggle--indeterminate" type="checkbox" id="__nav_7" >
<label class="md-nav__link" for="__nav_7" id="__nav_7_label" tabindex="0">
<span class="md-ellipsis">
About
</span>
<span class="md-nav__icon md-icon"></span>
</label>
<nav class="md-nav" data-md-level="1" aria-labelledby="__nav_7_label" aria-expanded="false">
<label class="md-nav__title" for="__nav_7">
<span class="md-nav__icon md-icon"></span>
About
</label>
<ul class="md-nav__list" data-md-scrollfix>
<li class="md-nav__item">
<a href="../LICENSE/" class="md-nav__link">
<span class="md-ellipsis">
License
</span>
</a>
</li>
<li class="md-nav__item">
<a href="../CHANGELOG/" class="md-nav__link">
<span class="md-ellipsis">
Changelog
</span>
</a>
</li>
</ul>
</nav>
</li>
</ul>
</nav>
</div>
</div>
</div>
<div class="md-sidebar md-sidebar--secondary" data-md-component="sidebar" data-md-type="toc" >
<div class="md-sidebar__scrollwrap">
<div class="md-sidebar__inner">
<nav class="md-nav md-nav--secondary" aria-label="Table of contents">
<label class="md-nav__title" for="__toc">
<span class="md-nav__icon md-icon"></span>
Table of contents
</label>
<ul class="md-nav__list" data-md-component="toc" data-md-scrollfix>
<li class="md-nav__item">
<a href="#overview" class="md-nav__link">
<span class="md-ellipsis">
Overview
</span>
</a>
</li>
<li class="md-nav__item">
<a href="#key-benefits" class="md-nav__link">
<span class="md-ellipsis">
Key Benefits
</span>
</a>
</li>
<li class="md-nav__item">
<a href="#architecture-components" class="md-nav__link">
<span class="md-ellipsis">
Architecture Components
</span>
</a>
<nav class="md-nav" aria-label="Architecture Components">
<ul class="md-nav__list">
<li class="md-nav__item">
<a href="#1-proxy-server" class="md-nav__link">
<span class="md-ellipsis">
1. Proxy Server
</span>
</a>
</li>
<li class="md-nav__item">
<a href="#2-local-llms" class="md-nav__link">
<span class="md-ellipsis">
2. Local LLMs
</span>
</a>
</li>
<li class="md-nav__item">
<a href="#3-mcp-tool-chain" class="md-nav__link">
<span class="md-ellipsis">
3. MCP Tool Chain
</span>
</a>
</li>
<li class="md-nav__item">
<a href="#4-validation-layer" class="md-nav__link">
<span class="md-ellipsis">
4. Validation Layer
</span>
</a>
</li>
</ul>
</nav>
</li>
<li class="md-nav__item">
<a href="#detailed-flow-description" class="md-nav__link">
<span class="md-ellipsis">
Detailed Flow Description
</span>
</a>
<nav class="md-nav" aria-label="Detailed Flow Description">
<ul class="md-nav__list">
<li class="md-nav__item">
<a href="#phase-1-initial-intake" class="md-nav__link">
<span class="md-ellipsis">
Phase 1: Initial Intake
</span>
</a>
</li>
<li class="md-nav__item">
<a href="#phase-2-intelligent-routing" class="md-nav__link">
<span class="md-ellipsis">
Phase 2: Intelligent Routing
</span>
</a>
</li>
<li class="md-nav__item">
<a href="#phase-3-prompt-optimization" class="md-nav__link">
<span class="md-ellipsis">
Phase 3: Prompt Optimization
</span>
</a>
</li>
<li class="md-nav__item">
<a href="#phase-4-validation-refinement" class="md-nav__link">
<span class="md-ellipsis">
Phase 4: Validation & Refinement
</span>
</a>
</li>
<li class="md-nav__item">
<a href="#phase-5-mcp-tool-chain-execution" class="md-nav__link">
<span class="md-ellipsis">
Phase 5: MCP Tool Chain Execution
</span>
</a>
</li>
<li class="md-nav__item">
<a href="#phase-6-final-assembly" class="md-nav__link">
<span class="md-ellipsis">
Phase 6: Final Assembly
</span>
</a>
</li>
<li class="md-nav__item">
<a href="#phase-7-claude-execution" class="md-nav__link">
<span class="md-ellipsis">
Phase 7: Claude Execution
</span>
</a>
</li>
</ul>
</nav>
</li>
<li class="md-nav__item">
<a href="#sequence-diagram" class="md-nav__link">
<span class="md-ellipsis">
Sequence Diagram
</span>
</a>
</li>
<li class="md-nav__item">
<a href="#configuration-options" class="md-nav__link">
<span class="md-ellipsis">
Configuration Options
</span>
</a>
<nav class="md-nav" aria-label="Configuration Options">
<ul class="md-nav__list">
<li class="md-nav__item">
<a href="#complexity-thresholds" class="md-nav__link">
<span class="md-ellipsis">
Complexity Thresholds
</span>
</a>
</li>
</ul>
</nav>
</li>
</ul>
</nav>
</div>
</div>
</div>
<div class="md-content" data-md-component="content">
<article class="md-content__inner md-typeset">
<a href="https://github.com/prefrontal-systems/cortexgraph/edit/main/docs/prompt_optimization_flow.md" title="Edit this page" class="md-content__button md-icon" rel="edit">
<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M10 20H6V4h7v5h5v3.1l2-2V8l-6-6H6c-1.1 0-2 .9-2 2v16c0 1.1.9 2 2 2h4zm10.2-7c.1 0 .3.1.4.2l1.3 1.3c.2.2.2.6 0 .8l-1 1-2.1-2.1 1-1c.1-.1.2-.2.4-.2m0 3.9L14.1 23H12v-2.1l6.1-6.1z"/></svg>
</a>
<h1 id="prompt-optimization-flow">Prompt Optimization Flow<a class="headerlink" href="#prompt-optimization-flow" title="Permanent link">¶</a></h1>
<h2 id="overview">Overview<a class="headerlink" href="#overview" title="Permanent link">¶</a></h2>
<p>This document describes a sophisticated prompt optimization architecture that intercepts, analyzes, enhances, and validates user prompts before they reach Claude. The system uses a multi-stage pipeline involving local LLMs, MCP tool chains, knowledge graph integration, and cloud-based optimization to maximize prompt quality while minimizing API costs.</p>
<h2 id="key-benefits">Key Benefits<a class="headerlink" href="#key-benefits" title="Permanent link">¶</a></h2>
<ul>
<li><strong>Zero Initial API Cost</strong>: All optimization happens before hitting paid Claude API endpoints</li>
<li><strong>Intelligent Complexity Routing</strong>: Simple prompts bypass optimization for speed; complex prompts get full treatment</li>
<li><strong>Knowledge Graph Integration</strong>: Automatically enriches prompts with relevant context from CortexGraph</li>
<li><strong>Multi-Model Validation</strong>: Cross-validates optimizations using multiple LLMs to ensure quality</li>
<li><strong>Flexible Architecture</strong>: Local LLMs can be swapped with cloud providers as needed</li>
<li><strong>Metadata Enrichment</strong>: Adds confidence scores, similarity metrics, and processing metadata to prompts</li>
</ul>
<h2 id="architecture-components">Architecture Components<a class="headerlink" href="#architecture-components" title="Permanent link">¶</a></h2>
<h3 id="1-proxy-server">1. <strong>Proxy Server</strong><a class="headerlink" href="#1-proxy-server" title="Permanent link">¶</a></h3>
<ul>
<li>Central orchestration layer</li>
<li>Handles routing decisions based on complexity</li>
<li>Manages communication between all components</li>
<li>Tracks confidence/similarity thresholds</li>
</ul>
<h3 id="2-local-llms">2. <strong>Local LLMs</strong><a class="headerlink" href="#2-local-llms" title="Permanent link">¶</a></h3>
<ul>
<li>Primary: Prompt optimization and tagging</li>
<li>Validation: Multiple instances for cross-validation</li>
<li>Can be replaced with cloud providers (OpenAI, Anthropic, etc.)</li>
</ul>
<h3 id="3-mcp-tool-chain">3. <strong>MCP Tool Chain</strong><a class="headerlink" href="#3-mcp-tool-chain" title="Permanent link">¶</a></h3>
<ul>
<li><strong>CortexGraph</strong>: Knowledge graph for context retrieval</li>
<li><strong>STOPPER</strong>: Process control and validation</li>
<li><strong>Custom Tools</strong>: User-defined extensions</li>
<li><strong>Gemini Optimizer</strong>: Large context window for final assembly</li>
</ul>
<h3 id="4-validation-layer">4. <strong>Validation Layer</strong><a class="headerlink" href="#4-validation-layer" title="Permanent link">¶</a></h3>
<ul>
<li>Semantic similarity checks</li>
<li>Confidence scoring</li>
<li>Iterative refinement below thresholds</li>
</ul>
<h2 id="detailed-flow-description">Detailed Flow Description<a class="headerlink" href="#detailed-flow-description" title="Permanent link">¶</a></h2>
<h3 id="phase-1-initial-intake">Phase 1: Initial Intake<a class="headerlink" href="#phase-1-initial-intake" title="Permanent link">¶</a></h3>
<ol>
<li><strong>User Input</strong>: User enters prompt in Claude Code interface</li>
<li><strong>Proxy Intercept</strong>: Proxy captures the prompt before it reaches Claude</li>
<li><strong>Complexity Analysis</strong>: NLP-based complexity rating determines routing strategy</li>
</ol>
<h3 id="phase-2-intelligent-routing">Phase 2: Intelligent Routing<a class="headerlink" href="#phase-2-intelligent-routing" title="Permanent link">¶</a></h3>
<ol>
<li><strong>Simple Path</strong> (Low Complexity):</li>
<li>Proxy applies basic formatting rules</li>
<li>Routes directly to Claude with minimal processing</li>
<li>
<p>Optimizes for speed and reduces overhead</p>
</li>
<li>
<p><strong>Complex Path</strong> (High Complexity):</p>
</li>
<li>Triggers full optimization pipeline</li>
<li>Proceeds to Phase 3</li>
</ol>
<h3 id="phase-3-prompt-optimization">Phase 3: Prompt Optimization<a class="headerlink" href="#phase-3-prompt-optimization" title="Permanent link">¶</a></h3>
<ol>
<li><strong>Local LLM Processing</strong>:</li>
<li>Adds semantic tags to categorize intent</li>
<li>Restructures prompt for optimal Claude comprehension</li>
<li>Formats according to Claude best practices</li>
<li>Extracts key entities and concepts</li>
</ol>
<h3 id="phase-4-validation-refinement">Phase 4: Validation & Refinement<a class="headerlink" href="#phase-4-validation-refinement" title="Permanent link">¶</a></h3>
<ol>
<li><strong>Multi-Model Validation</strong>:</li>
<li>Routes optimized prompt to 2-n additional local LLMs</li>
<li>Each validator scores the optimization independently</li>
<li>Can use semantic similarity algorithms instead of LLMs</li>
<li>
<p>Calculates confidence and similarity metrics</p>
</li>
<li>
<p><strong>Threshold Check</strong>:</p>
</li>
<li>If scores meet threshold: Proceed to Phase 5</li>
<li>If scores below threshold: Return to Phase 3 for reprocessing</li>
<li>
<p>Prevents low-quality optimizations from proceeding</p>
</li>
<li>
<p><strong>Tool Recommendation</strong>:</p>
</li>
<li>Proxy receives validated prompt with metadata</li>
<li>System suggests relevant MCP tools for the query</li>
</ol>
<h3 id="phase-5-mcp-tool-chain-execution">Phase 5: MCP Tool Chain Execution<a class="headerlink" href="#phase-5-mcp-tool-chain-execution" title="Permanent link">¶</a></h3>
<ol>
<li>
<p><strong>CortexGraph Search</strong>:</p>
<ul>
<li>Searches knowledge graph for related concepts</li>
<li>Retrieves relevant memories and context</li>
<li>Returns similarity-scored results</li>
</ul>
</li>
<li>
<p><strong>STOPPER Validation</strong>:</p>
<ul>
<li>Process control checks</li>
<li>Safety and constraint validation</li>
<li>Prevents out-of-scope operations</li>
</ul>
</li>
<li>
<p><strong>Additional Tools</strong>:</p>
<ul>
<li>Routes to n other tools based on user preferences</li>
<li>Each tool contributes specialized context</li>
<li>Tools run in parallel for efficiency</li>
</ul>
</li>
</ol>
<h3 id="phase-6-final-assembly">Phase 6: Final Assembly<a class="headerlink" href="#phase-6-final-assembly" title="Permanent link">¶</a></h3>
<ol>
<li>
<p><strong>Gemini Optimization</strong>:</p>
<ul>
<li>Combines original prompt + optimizations + tool outputs</li>
<li>Leverages Gemini's large context window (2M tokens)</li>
<li>Uses generous free tier for cost optimization</li>
<li>Assembles coherent final prompt</li>
</ul>
</li>
<li>
<p><strong>Quality Assurance</strong>:</p>
<ul>
<li>Compares input to assembled output</li>
<li>Generates similarity score (drift detection)</li>
<li>Calculates final confidence rating</li>
<li>Appends metadata to prompt</li>
</ul>
</li>
</ol>
<h3 id="phase-7-claude-execution">Phase 7: Claude Execution<a class="headerlink" href="#phase-7-claude-execution" title="Permanent link">¶</a></h3>
<ol>
<li>
<p><strong>Final Prompt Delivery</strong>:</p>
<ul>
<li>Proxy sends optimized prompt to Claude</li>
<li><strong>First API cost incurred at this step</strong></li>
<li>Prompt includes:</li>
<li>Original user intent (preserved)</li>
<li>Optimization tags and structure</li>
<li>Knowledge graph context</li>
<li>Tool outputs and recommendations</li>
<li>Confidence/similarity metadata</li>
<li>Processing history</li>
</ul>
</li>
<li>
<p><strong>Normal Operation</strong>:</p>
<ul>
<li>Claude processes the enriched prompt</li>
<li>Claude Code continues standard workflow</li>
<li>User receives high-quality response</li>
</ul>
</li>
</ol>
<h2 id="sequence-diagram">Sequence Diagram<a class="headerlink" href="#sequence-diagram" title="Permanent link">¶</a></h2>
<p>```mermaid
sequenceDiagram
actor User
participant Claude Code Interface
participant Proxy
participant NLP Complexity Analyzer
participant Local LLM (Optimizer)
participant Local LLM 2 (Validator)
participant Local LLM N (Validator)
participant Semantic Similarity Engine
participant MCP Chain
participant CortexGraph
participant STOPPER
participant Custom Tools
participant Gemini
participant Claude API</p>
<div class="highlight"><pre><span></span><code>%% Phase 1: Initial Intake
User->>Claude Code Interface: Enter prompt
Claude Code Interface->>Proxy: Forward prompt
Proxy->>NLP Complexity Analyzer: Analyze complexity
NLP Complexity Analyzer-->>Proxy: Complexity rating
%% Phase 2: Routing Decision
alt Low Complexity (Simple Prompt)
Proxy->>Proxy: Apply basic rules
Proxy->>Claude API: Route directly to Claude
Note over Proxy,Claude API: Fast path for simple queries
else High Complexity (Complex Prompt)
Note over Proxy: Trigger full optimization pipeline
%% Phase 3: Optimization
Proxy->>Local LLM (Optimizer): Optimize prompt
Note over Local LLM (Optimizer): - Add semantic tags<br/>- Format for Claude<br/>- Extract entities<br/>- Restructure query
Local LLM (Optimizer)-->>Proxy: Optimized prompt v1
%% Phase 4: Validation Loop
rect rgb(240, 240, 240)
Note over Proxy,Semantic Similarity Engine: Validation & Refinement Loop
par Parallel Validation
Proxy->>Local LLM 2 (Validator): Validate optimization
Proxy->>Local LLM N (Validator): Validate optimization
Proxy->>Semantic Similarity Engine: Check semantic similarity
end
Local LLM 2 (Validator)-->>Proxy: Confidence score 2
Local LLM N (Validator)-->>Proxy: Confidence score N
Semantic Similarity Engine-->>Proxy: Similarity score
Proxy->>Proxy: Aggregate scores
alt Below Confidence/Similarity Threshold
Note over Proxy,Local LLM (Optimizer): Quality check failed
Proxy->>Local LLM (Optimizer): Reprocess with feedback
Local LLM (Optimizer)-->>Proxy: Optimized prompt v2
Note over Proxy: Loop until threshold met
else Above Threshold
Note over Proxy: Quality validated, proceed
end
end
Proxy->>Proxy: Append recommendation metadata
%% Phase 5: MCP Tool Chain
Proxy->>MCP Chain: Route validated prompt + metadata
rect rgb(230, 245, 255)
Note over MCP Chain,Custom Tools: MCP Tool Execution (Parallel)
par Tool Execution
MCP Chain->>CortexGraph: Search knowledge graph
MCP Chain->>STOPPER: Validate constraints
MCP Chain->>Custom Tools: Execute user-defined tools
end
CortexGraph-->>MCP Chain: Context + memories (similarity scored)
STOPPER-->>MCP Chain: Validation results
Custom Tools-->>MCP Chain: Tool outputs
end
%% Phase 6: Final Assembly
MCP Chain->>Gemini: Assemble final prompt
Note over Gemini: - Combine all inputs<br/>- Optimize structure<br/>- 2M token context<br/>- Free tier usage
Gemini->>Gemini: Compare input vs output
Gemini->>Gemini: Calculate similarity & confidence
Gemini-->>MCP Chain: Final prompt + metadata
MCP Chain-->>Proxy: Return final prompt
%% Phase 7: Claude Execution
Note over Proxy,Claude API: 💰 First API cost incurred here
Proxy->>Claude API: Send final optimized prompt
Note over Claude API: Prompt includes:<br/>- Original intent<br/>- Optimizations<br/>- Knowledge graph context<br/>- Tool outputs<br/>- Metadata
end
%% Normal Operation
Claude API-->>Claude Code Interface: Process request
Claude Code Interface-->>User: Return response
Note over User,Claude Code Interface: Claude Code continues as normal
</code></pre></div>
<p>```</p>
<h2 id="configuration-options">Configuration Options<a class="headerlink" href="#configuration-options" title="Permanent link">¶</a></h2>
<h3 id="complexity-thresholds">Complexity Thresholds<a class="headerlink" href="#complexity-thresholds" title="Permanent link">¶</a></h3>
<p>```python</p>
<h1 id="proxy-configuration">Proxy configuration<a class="headerlink" href="#proxy-configuration" title="Permanent link">¶</a></h1>
<h1 id="prompts-with-complexity-complex_prompt_threshold-follow-the-complex-path-otherwise-the-simple-path-is-used">Prompts with complexity > COMPLEX_PROMPT_THRESHOLD follow the complex path, otherwise the simple path is used.<a class="headerlink" href="#prompts-with-complexity-complex_prompt_threshold-follow-the-complex-path-otherwise-the-simple-path-is-used" title="Permanent link">¶</a></h1>
<p>COMPLEX_PROMPT_THRESHOLD = 0.4
```</p>
<h3 id="validation-settings">Validation Settings<a class="headerlink" href="#validation-settings" title="Permanent link">¶</a></h3>
<p>```python</p>
<h1 id="validation-thresholds">Validation thresholds<a class="headerlink" href="#validation-thresholds" title="Permanent link">¶</a></h1>
<p>CONFIDENCE_THRESHOLD = 0.75 # Minimum confidence to proceed
SIMILARITY_THRESHOLD = 0.80 # Minimum semantic similarity
MAX_REFINEMENT_ITERATIONS = 3 # Prevent infinite loops
```</p>
<h3 id="model-selection">Model Selection<a class="headerlink" href="#model-selection" title="Permanent link">¶</a></h3>
<p>```python</p>
<h1 id="local-llms-can-be-replaced-with-cloud-providers">Local LLMs (can be replaced with cloud providers)<a class="headerlink" href="#local-llms-can-be-replaced-with-cloud-providers" title="Permanent link">¶</a></h1>
<p>OPTIMIZER_MODEL = "llama-3.1-70b" # Primary optimizer
VALIDATOR_MODELS = [ # Validation ensemble
"mixtral-8x7b",
"qwen-2.5-72b",
"deepseek-v2"
]</p>
<h1 id="example-using-cloud-providers-alternative-to-local">Example using cloud providers (alternative to local)<a class="headerlink" href="#example-using-cloud-providers-alternative-to-local" title="Permanent link">¶</a></h1>
<h1 id="optimizer_model-openaigpt-4">OPTIMIZER_MODEL = "openai:gpt-4"<a class="headerlink" href="#optimizer_model-openaigpt-4" title="Permanent link">¶</a></h1>
<h1 id="validator_models-anthropicclaude-3-opus-openaigpt-4">VALIDATOR_MODELS = ["anthropic:claude-3-opus", "openai:gpt-4"]<a class="headerlink" href="#validator_models-anthropicclaude-3-opus-openaigpt-4" title="Permanent link">¶</a></h1>
<p>```</p>
<h3 id="mcp-tools">MCP Tools<a class="headerlink" href="#mcp-tools" title="Permanent link">¶</a></h3>
<p>```python</p>
<h1 id="tool-chain-configuration">Tool chain configuration<a class="headerlink" href="#tool-chain-configuration" title="Permanent link">¶</a></h1>
<p>MCP_TOOLS = {
"cortex_graph": {
"enabled": True,
"similarity_threshold": 0.7,
"max_results": 10
},
"stopper": {
"enabled": True,
"strict_mode": False
},
"custom": {
"user_preferences": True,
"context_retrieval": True
}
}
```</p>
<h3 id="gemini-settings">Gemini Settings<a class="headerlink" href="#gemini-settings" title="Permanent link">¶</a></h3>
<p>```python</p>
<h1 id="final-assembly-configuration">Final assembly configuration<a class="headerlink" href="#final-assembly-configuration" title="Permanent link">¶</a></h1>
<p>GEMINI_MODEL = "gemini-2.0-flash-exp" # Free tier, large context
GEMINI_MAX_TOKENS = 2000000 # 2M token context window
GEMINI_TEMPERATURE = 0.3 # Consistent assembly
```</p>
<h2 id="performance-characteristics">Performance Characteristics<a class="headerlink" href="#performance-characteristics" title="Permanent link">¶</a></h2>
<h3 id="latency-profile">Latency Profile<a class="headerlink" href="#latency-profile" title="Permanent link">¶</a></h3>
<table>
<thead>
<tr>
<th>Stage</th>
<th>Estimated Time</th>
<th>Notes</th>
</tr>
</thead>
<tbody>
<tr>
<td>Complexity Analysis</td>
<td>10-50ms</td>
<td>Fast NLP classification</td>
</tr>
<tr>
<td>Simple Path (total)</td>
<td>50-100ms</td>
<td>Minimal processing overhead</td>
</tr>
<tr>
<td>Optimization</td>
<td>200-500ms</td>
<td>Local LLM inference</td>
</tr>
<tr>
<td>Validation</td>
<td>150-300ms</td>
<td>Parallel execution</td>
</tr>
<tr>
<td>MCP Tool Chain</td>
<td>100-400ms</td>
<td>Depends on tool complexity</td>
</tr>
<tr>
<td>Gemini Assembly</td>
<td>300-800ms</td>
<td>Large context processing</td>
</tr>
<tr>
<td><strong>Complex Path (total)</strong></td>
<td><strong>1-3 seconds</strong></td>
<td>Full pipeline</td>
</tr>
</tbody>
</table>
<h3 id="cost-analysis">Cost Analysis<a class="headerlink" href="#cost-analysis" title="Permanent link">¶</a></h3>
<p><strong>Traditional Approach</strong> (direct to Claude):
- Every prompt hits Claude API immediately
- No optimization or context enrichment
- Cost: $X per request from first token</p>
<p><strong>Optimized Approach</strong> (this architecture):
- Local LLMs: Free (self-hosted) or cheap (cloud)
- Gemini: Leverages the generous free tier for final assembly
- Claude API: Only hit after full optimization
- Cost: $0 until Claude execution, then same $X but better results</p>
<p><strong>Net Effect</strong>:
- Same Claude API cost per request
- Significantly better prompt quality
- Higher success rate (fewer retries needed)
- Lower total cost due to reduced iterations</p>
<h2 id="implementation-considerations">Implementation Considerations<a class="headerlink" href="#implementation-considerations" title="Permanent link">¶</a></h2>
<h3 id="1-local-llm-requirements">1. <strong>Local LLM Requirements</strong><a class="headerlink" href="#1-local-llm-requirements" title="Permanent link">¶</a></h3>
<ul>
<li>GPU: RTX 4090 or better for 70B models</li>
<li>RAM: 64GB+ recommended</li>
<li>Alternative: Use cloud inference APIs (Groq, Together.ai, OpenRouter)</li>
</ul>
<h3 id="2-proxy-server">2. <strong>Proxy Server</strong><a class="headerlink" href="#2-proxy-server" title="Permanent link">¶</a></h3>
<ul>
<li>Needs to be MCP-compatible</li>
<li>Should support WebSocket for streaming</li>
<li>Must handle concurrent validation requests</li>
</ul>
<h3 id="3-knowledge-graph-integration">3. <strong>Knowledge Graph Integration</strong><a class="headerlink" href="#3-knowledge-graph-integration" title="Permanent link">¶</a></h3>
<ul>
<li>CortexGraph needs to be populated with relevant data</li>
<li>Index must be kept up-to-date</li>
<li>Consider using CortexGraph for temporal memory</li>
</ul>
<h3 id="4-error-handling">4. <strong>Error Handling</strong><a class="headerlink" href="#4-error-handling" title="Permanent link">¶</a></h3>
<ul>
<li>Fallback to simple path if optimization fails</li>
<li>Timeout protection (max 5s total processing)</li>
<li>Graceful degradation if tools unavailable</li>
</ul>
<h3 id="5-monitoring-observability">5. <strong>Monitoring & Observability</strong><a class="headerlink" href="#5-monitoring-observability" title="Permanent link">¶</a></h3>
<ul>
<li>Track optimization success rates</li>
<li>Monitor confidence/similarity distributions</li>
<li>Log processing times for each stage</li>
<li>A/B test optimized vs non-optimized prompts</li>
</ul>
<h2 id="future-enhancements">Future Enhancements<a class="headerlink" href="#future-enhancements" title="Permanent link">¶</a></h2>
<ol>
<li><strong>Adaptive Thresholds</strong>: Learn optimal confidence/similarity thresholds per user</li>
<li><strong>Caching Layer</strong>: Cache optimizations for similar prompts</li>
<li><strong>User Feedback Loop</strong>: Incorporate user ratings to improve optimization</li>
<li><strong>Model Selection</strong>: Automatically choose best LLM based on prompt type</li>
<li><strong>Streaming Optimization</strong>: Stream partial results during processing</li>
<li><strong>Cost Tracking</strong>: Detailed cost accounting per stage</li>
<li><strong>A/B Testing Framework</strong>: Compare different optimization strategies</li>
</ol>
<h2 id="security-considerations">Security Considerations<a class="headerlink" href="#security-considerations" title="Permanent link">¶</a></h2>
<ul>
<li><strong>Prompt Injection</strong>: Validate all optimized prompts for injection attempts</li>
<li><strong>Data Privacy</strong>: Local LLMs keep sensitive data on-premise</li>
<li><strong>Rate Limiting</strong>: Prevent abuse of free tier services</li>
<li><strong>Access Control</strong>: Authenticate proxy requests</li>
<li><strong>Audit Trail</strong>: Log all prompt transformations</li>
</ul>
<h2 id="related-documentation">Related Documentation<a class="headerlink" href="#related-documentation" title="Permanent link">¶</a></h2>
<ul>
<li><a href="../architecture/">CortexGraph Architecture</a> - Integration with temporal memory</li>
<li><a href="../graph_features/">CortexGraph Documentation</a> - Knowledge graph features</li>
<li><a href="https://github.com/modelcontextprotocol/specification">MCP Specification</a> - Tool protocol details</li>
<li><a href="../prompt_injection/">Prompt Injection Prevention</a> - Security best practices</li>
</ul>
<h2 id="example-workflow">Example Workflow<a class="headerlink" href="#example-workflow" title="Permanent link">¶</a></h2>
<h3 id="input-prompt">Input Prompt<a class="headerlink" href="#input-prompt" title="Permanent link">¶</a></h3>
<p>```
"Help me write a Python function to process user data"
```</p>
<h3 id="after-optimization">After Optimization<a class="headerlink" href="#after-optimization" title="Permanent link">¶</a></h3>
<p>```markdown</p>
<h2 id="task-python-function-development">Task: Python Function Development<a class="headerlink" href="#task-python-function-development" title="Permanent link">¶</a></h2>
<p><strong>User Intent</strong>: Create data processing function</p>
<p><strong>Context</strong> (from CortexGraph):
- User prefers type hints (from memory: 2025-10-15)
- Uses pytest for testing (from memory: 2025-10-20)
- Prefers dataclasses over dicts (from memory: 2025-10-12)</p>
<p><strong>Requirements</strong>:
1. Function should process user data
2. Follow user's Python style preferences
3. Include type hints and docstrings
4. Consider testing approach</p>
<p><strong>Metadata</strong>:
- Confidence: 0.87
- Similarity: 0.92
- Optimization iterations: 1
- Tools used: CortexGraph, STOPPER
- Processing time: 1.2s
```</p>
<h3 id="result">Result<a class="headerlink" href="#result" title="Permanent link">¶</a></h3>
<p>Claude receives a rich, contextualized prompt that produces higher-quality output on the first try, reducing the need for follow-up iterations.</p>
<hr />
<p><strong>Built with</strong> <a href="https://claude.com/claude-code">Claude Code</a> 🤖</p>
</article>
</div>
<script>var target=document.getElementById(location.hash.slice(1));target&&target.name&&(target.checked=target.name.startsWith("__tabbed_"))</script>
</div>
<button type="button" class="md-top md-icon" data-md-component="top" hidden>
<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M13 20h-2V8l-5.5 5.5-1.42-1.42L12 4.16l7.92 7.92-1.42 1.42L13 8z"/></svg>
Back to top
</button>
</main>
<footer class="md-footer">
<div class="md-footer-meta md-typeset">
<div class="md-footer-meta__inner md-grid">
<div class="md-copyright">
Made with
<a href="https://squidfunk.github.io/mkdocs-material/" target="_blank" rel="noopener">
Material for MkDocs
</a>
</div>
<div class="md-social">
<a href="https://github.com/prefrontal-systems/cortexgraph" target="_blank" rel="noopener" title="github.com" class="md-social__link">
<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 512 512"><!--! Font Awesome Free 7.1.0 by @fontawesome - https://fontawesome.com License - https://fontawesome.com/license/free (Icons: CC BY 4.0, Fonts: SIL OFL 1.1, Code: MIT License) Copyright 2025 Fonticons, Inc.--><path d="M173.9 397.4c0 2-2.3 3.6-5.2 3.6-3.3.3-5.6-1.3-5.6-3.6 0-2 2.3-3.6 5.2-3.6 3-.3 5.6 1.3 5.6 3.6m-31.1-4.5c-.7 2 1.3 4.3 4.3 4.9 2.6 1 5.6 0 6.2-2s-1.3-4.3-4.3-5.2c-2.6-.7-5.5.3-6.2 2.3m44.2-1.7c-2.9.7-4.9 2.6-4.6 4.9.3 2 2.9 3.3 5.9 2.6 2.9-.7 4.9-2.6 4.6-4.6-.3-1.9-3-3.2-5.9-2.9M252.8 8C114.1 8 8 113.3 8 252c0 110.9 69.8 205.8 169.5 239.2 12.8 2.3 17.3-5.6 17.3-12.1 0-6.2-.3-40.4-.3-61.4 0 0-70 15-84.7-29.8 0 0-11.4-29.1-27.8-36.6 0 0-22.9-15.7 1.6-15.4 0 0 24.9 2 38.6 25.8 21.9 38.6 58.6 27.5 72.9 20.9 2.3-16 8.8-27.1 16-33.7-55.9-6.2-112.3-14.3-112.3-110.5 0-27.5 7.6-41.3 23.6-58.9-2.6-6.5-11.1-33.3 2.6-67.9 20.9-6.5 69 27 69 27 20-5.6 41.5-8.5 62.8-8.5s42.8 2.9 62.8 8.5c0 0 48.1-33.6 69-27 13.7 34.7 5.2 61.4 2.6 67.9 16 17.7 25.8 31.5 25.8 58.9 0 96.5-58.9 104.2-114.8 110.5 9.2 7.9 17 22.9 17 46.4 0 33.7-.3 75.4-.3 83.6 0 6.5 4.6 14.4 17.3 12.1C436.2 457.8 504 362.9 504 252 504 113.3 391.5 8 252.8 8M105.2 352.9c-1.3 1-1 3.3.7 5.2 1.6 1.6 3.9 2.3 5.2 1 1.3-1 1-3.3-.7-5.2-1.6-1.6-3.9-2.3-5.2-1m-10.8-8.1c-.7 1.3.3 2.9 2.3 3.9 1.6 1 3.6.7 4.3-.7.7-1.3-.3-2.9-2.3-3.9-2-.6-3.6-.3-4.3.7m32.4 35.6c-1.6 1.3-1 4.3 1.3 6.2 2.3 2.3 5.2 2.6 6.5 1 1.3-1.3.7-4.3-1.3-6.2-2.2-2.3-5.2-2.6-6.5-1m-11.4-14.7c-1.6 1-1.6 3.6 0 5.9s4.3 3.3 5.6 2.3c1.6-1.3 1.6-3.9 0-6.2-1.4-2.3-4-3.3-5.6-2"/></svg>
</a>
<a href="https://github.com/prefrontal-systems/cortexgraph/discussions" target="_blank" rel="noopener" title="github.com" class="md-social__link">
<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 576 512"><!--! Font Awesome Free 7.1.0 by @fontawesome - https://fontawesome.com License - https://fontawesome.com/license/free (Icons: CC BY 4.0, Fonts: SIL OFL 1.1, Code: MIT License) Copyright 2025 Fonticons, Inc.--><path d="M384 144c0 97.2-86 176-192 176-26.7 0-52.1-5-75.2-14l-81.6 43.2c-9.3 4.9-20.7 3.2-28.2-4.2s-9.2-18.9-4.2-28.2l35.6-67.2C14.3 220.2 0 183.6 0 144 0 46.8 86-32 192-32s192 78.8 192 176m0 368c-94.1 0-172.4-62.1-188.8-144 120-1.5 224.3-86.9 235.8-202.7 83.3 19.2 145 88.3 145 170.7 0 39.6-14.3 76.2-38.4 105.6l35.6 67.2c4.9 9.3 3.2 20.7-4.2 28.2s-18.9 9.2-28.2 4.2L459.2 498c-23.1 9-48.5 14-75.2 14"/></svg>
</a>
</div>
</div>
</div>
</footer>
</div>
<div class="md-dialog" data-md-component="dialog">
<div class="md-dialog__inner md-typeset"></div>
</div>
<script id="__config" type="application/json">{"annotate": null, "base": "..", "features": ["navigation.instant", "navigation.tracking", "navigation.tabs", "navigation.sections", "navigation.expand", "navigation.top", "search.suggest", "search.highlight", "content.code.copy", "content.code.annotate", "content.action.edit"], "search": "../assets/javascripts/workers/search.7a47a382.min.js", "tags": null, "translations": {"clipboard.copied": "Copied to clipboard", "clipboard.copy": "Copy to clipboard", "search.result.more.one": "1 more on this page", "search.result.more.other": "# more on this page", "search.result.none": "No matching documents", "search.result.one": "1 matching document", "search.result.other": "# matching documents", "search.result.placeholder": "Type to start searching", "search.result.term.missing": "Missing", "select.version": "Select version"}, "version": null}</script>
<script src="../assets/javascripts/bundle.e71a0d61.min.js"></script>
</body>
</html>
Latest Blog Posts
- MCP Moves to the Linux Foundation: Neutral Stewardship for Agentic InfrastructureBy Om-Shree-0709 on .mcpanthropicLinux Foundation
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/prefrontalsys/mnemex'
If you have feedback or need assistance with the MCP directory API, please join our Discord server