Skip to content

The one and only agent harness for complex codebases. Project memory, planning, execution, and verified completion inside Codex.

fix(web): position LazyCodex as codebase harness

YeonGyu-Kim98bdf9ce7f69
diff --git a/README.md b/README.md
index 417accbc..4221748c 100644
--- a/README.md
+++ b/README.md
@@ -3,8 +3,8 @@
<h1>LazyCodex</h1>
- <p><strong>Codex for no-brainers.</strong><br />
- You don't need to think. Just prompt with <code>ultrawork</code>.</p>
+ <p><strong>The one and only agent harness for complex codebases.</strong><br />
+ Project memory, planning, execution, and verified completion inside Codex.</p>
<p>
<a href="https://github.com/code-yeongyu/lazycodex/stargazers">
@@ -54,9 +54,9 @@ Full documentation lives at [lazycodex.ai/docs](https://lazycodex.ai/docs).
## Use the built-in workflows
-LazyCodex should be judged by the features it actually installs. It is a thin
-Codex distribution for OmO: project memory, planning, execution, verified
-completion, skills, hooks, model routing, and diagnostics.
+LazyCodex should be judged by the features it actually installs. It is the
+Codex distribution for OmO's agent harness: project memory, planning,
+execution, verified completion, skills, hooks, model routing, and diagnostics.
### 1. `/init-deep` creates project memory
@@ -104,11 +104,11 @@ Start at [https://lazycodex.ai](https://lazycodex.ai).
## πŸ’€ What is this?
-**LazyCodex** is the **lazy way** to get [OmO (oh-my-openagent)](https://github.com/code-yeongyu/oh-my-openagent) up and running.
+**LazyCodex** packages [OmO (oh-my-openagent)](https://github.com/code-yeongyu/oh-my-openagent) as the Codex agent harness for complex codebases.
Think [LazyVim](https://github.com/LazyVim/LazyVim) for [lazy.nvim](https://github.com/folke/lazy.nvim), but for Codex.
-OmO is the best agent harness: discipline agents, parallel orchestration, multi-model routing, skills, hooks, and more. LazyCodex wraps it so you don't have to think about setup.
+OmO is the agent harness: discipline agents, parallel orchestration, multi-model routing, skills, hooks, and verified completion. LazyCodex packages that harness for Codex.
> _"LazyVim made Neovim usable for the rest of us. LazyCodex does the same for Codex."_
diff --git a/packages/web/DESIGN.md b/packages/web/DESIGN.md
index 7249c149..eed7af44 100644
--- a/packages/web/DESIGN.md
+++ b/packages/web/DESIGN.md
@@ -1,7 +1,7 @@
# LazyCodex Design System
## 1. Core Philosophy
-- **Codex-tone Luminous Backdrop**: The page uses a near-black canvas (`#0a0a0a`) with a centered, glowing teal card.
+- **Complex-Codebase Harness Tone**: The page uses a near-black canvas (`#0a0a0a`) with a centered, glowing teal card that presents LazyCodex as the Codex agent harness for serious repositories.
- **Card-in-Canvas Architecture**: The main content lives inside a 1200x630px card with a complex radial gradient, and the OpenGraph image mirrors that HTML card instead of using a separate visual language.
- **Typography**: Clean, geometric sans-serif (Geist Sans) for the wordmark and tagline, with monospace (Geist Mono) for eyebrows and footers.
@@ -53,7 +53,7 @@ Sisyphus Labs colors are used as accents across the landing page while the black
Where they appear:
- GitHub stars pill in the sticky header
- Command-card accents (borders, icons, hover states)
-- The `ultrawork` keyword glow inside the hero tagline
+- The `verified completion` keyword glow inside the hero tagline
- CTA and link accents (hover underlines, button fills)
- Footer co-brand strip
@@ -68,14 +68,14 @@ The landing page is a single scrollable document with the following sections, to
- "by Sisyphus Labs" co-brand
2. **Hero luminous card**
- - Eyebrow: "CODEX FOR NO-BRAINERS"
+ - Eyebrow: "AGENT HARNESS FOR COMPLEX CODEBASES"
- H1 wordmark
- - Tagline: "You don't need to ultrathink. / Just prompt {your prompt} ultrawork."
- - The `{your prompt}` placeholder uses Geist Mono and pulses subtly
+ - Tagline: "The one and only agent harness for complex codebases. / Project memory, planning, execution, and verified completion."
+ - The `planning,` placeholder uses Geist Mono and pulses subtly
3. **Install block**
- Copyable bash snippet: `npx lazycodex-ai install`
- - One-line note: "No global install. Always `npx`."
+ - One-line note: equivalent OmO install command
- Secondary line for the autonomous variant: `npx lazycodex-ai install --no-tui --codex-autonomous`
4. **Three command cards**
@@ -84,7 +84,7 @@ The landing page is a single scrollable document with the following sections, to
- `$start-work` β€” plan executor
- Each card has a monospace command header, a one-line description, and a syntax example
-5. **Built-in workflows + skill coverage**
+5. **Harness workflows + skill coverage**
- The existing three command cards remain the main pillars and appear before this guide
- Explains the installed LazyCodex workflows: `/init-deep`, `$ulw-plan`, `$start-work`, and `$ulw-loop`
- Calls out project memory, planning, execution, and verified completion
@@ -116,7 +116,7 @@ The landing page is a single scrollable document with the following sections, to
## 9. Image pipeline
-- **OpenGraph / Twitter card**: `app/opengraph-image.tsx` and `app/twitter-image.tsx` use `next/og` and inline the same black/teal tokens, gradient layers, wordmark, tagline, `{your prompt}` pill, URL, and install command seen in the HTML hero.
+- **OpenGraph / Twitter card**: `app/opengraph-image.tsx` and `app/twitter-image.tsx` use `next/og` and inline the same black/teal tokens, gradient layers, wordmark, harness tagline, planning pill, pillar chips, URL, and install command seen in the HTML hero.
- **Boulder art**: Generated via `imagegen` on a chroma background, keyed to alpha so it composites cleanly on the dark canvas without a visible bounding box
- **Delivery format**: Served as AVIF / WebP / PNG via a `<picture>` element using the `BrandImage` component
- **Dimensions**: Explicit `width` and `height` attributes on every image to prevent CLS
diff --git a/packages/web/app/docs/page.tsx b/packages/web/app/docs/page.tsx
index b03edc4e..cfac2e36 100644
--- a/packages/web/app/docs/page.tsx
+++ b/packages/web/app/docs/page.tsx
@@ -9,7 +9,7 @@ import "@/app/styles/docs.css";
export const metadata: Metadata = {
title: "Documentation",
description:
- "LazyCodex documentation β€” install, built-in workflows, skills, ultrawork mode, $ulw-loop, $ulw-plan, $start-work.",
+ "LazyCodex documentation for the OmO agent harness: install, project memory, skills, ultrawork mode, $ulw-loop, $ulw-plan, and $start-work.",
alternates: {
canonical: "/docs",
},
@@ -28,7 +28,7 @@ export default function DocsPage() {
<div className="docs-hero">
<h1 className="docs-hero-title">Documentation</h1>
<p className="docs-hero-tagline">
- Learn how to install, configure, and use LazyCodex for ultrawork and built-in workflows.
+ Learn how to install and run the OmO harness inside Codex for complex codebases.
</p>
</div>
<DocsShell sections={sections}>
diff --git a/packages/web/app/layout.tsx b/packages/web/app/layout.tsx
index 2bc9b1a5..15c5bf8d 100644
--- a/packages/web/app/layout.tsx
+++ b/packages/web/app/layout.tsx
@@ -5,14 +5,14 @@ import "./globals.css"
const SITE_URL = "https://lazycodex.ai"
const SITE_NAME = "LazyCodex"
-const TAGLINE = "Codex for no-brainers"
+const TAGLINE = "Codex agent harness for complex codebases"
const DESCRIPTION =
- "LazyCodex. You don't need to ultrathink. Just prompt with ultrawork and install the Codex agent harness in one command."
+ "LazyCodex installs the OmO agent harness inside Codex: project memory, planning, parallel agents, skills, hooks, routing, and verified completion for complex codebases."
export const viewport: Viewport = {
width: "device-width",
initialScale: 1,
- themeColor: "#2a2dbf",
+ themeColor: "#008385",
colorScheme: "dark",
}
@@ -33,6 +33,9 @@ export const metadata: Metadata = {
"ai agent",
"code agent",
"agent harness",
+ "complex codebases",
+ "verified completion",
+ "project memory",
"sisyphus",
"lazyvim",
],
diff --git a/packages/web/app/manifest.ts b/packages/web/app/manifest.ts
index 51ac11f8..6401d356 100644
--- a/packages/web/app/manifest.ts
+++ b/packages/web/app/manifest.ts
@@ -4,11 +4,11 @@ export default function manifest(): MetadataRoute.Manifest {
return {
name: "LazyCodex",
short_name: "LazyCodex",
- description: "Codex for no-brainers. Just prompt with ultrawork.",
+ description: "Agent harness for complex codebases inside Codex.",
start_url: "/",
display: "standalone",
- background_color: "#14154d",
- theme_color: "#2a2dbf",
+ background_color: "#0a0a0a",
+ theme_color: "#008385",
// No icons array: the browser favicon comes from the app/icon.svg file
// convention; duplicating it here triggers a second eager favicon fetch
// that lands on the Lantern LCP critical path. PWA installability isn't a
diff --git a/packages/web/app/opengraph-image.tsx b/packages/web/app/opengraph-image.tsx
index 91d81b6a..9318a543 100644
--- a/packages/web/app/opengraph-image.tsx
+++ b/packages/web/app/opengraph-image.tsx
@@ -4,7 +4,7 @@ import { OgBrandMark } from "./og-brand-mark"
import { OG_GRADIENTS, OG_PALETTE, OG_SIZE } from "./og-image-theme"
export const runtime = "nodejs"
-export const alt = "LazyCodex - Codex for no-brainers."
+export const alt = "LazyCodex - agent harness for complex codebases."
export const size = OG_SIZE
export const contentType = "image/png"
@@ -107,12 +107,12 @@ export default function OgImage(): ImageResponse {
position: "relative",
display: "flex",
flexDirection: "column",
- alignItems: "center",
+ alignItems: "flex-start",
justifyContent: "center",
width: "100%",
height: "100%",
- padding: "70px 72px 62px",
- textAlign: "center",
+ padding: "70px 86px 62px",
+ textAlign: "left",
}}
>
<div
@@ -131,12 +131,12 @@ export default function OgImage(): ImageResponse {
</div>
<div
style={{
- marginTop: "28px",
+ marginTop: "18px",
color: palette.textPrimary,
- fontSize: "142px",
+ fontSize: "104px",
fontWeight: 500,
- letterSpacing: "-4.3px",
- lineHeight: 0.98,
+ letterSpacing: "-3.1px",
+ lineHeight: 1.1,
}}
>
{SITE_CONFIG.wordmark}
@@ -145,18 +145,19 @@ export default function OgImage(): ImageResponse {
style={{
display: "flex",
flexDirection: "column",
- alignItems: "center",
- marginTop: "30px",
- gap: "2px",
+ alignItems: "flex-start",
+ marginTop: "56px",
+ gap: "8px",
color: palette.textMuted,
- fontSize: "33px",
+ fontSize: "36px",
fontWeight: 400,
letterSpacing: "-0.17px",
- lineHeight: 1.35,
+ lineHeight: 1.2,
+ width: "880px",
}}
>
<div>{SITE_CONFIG.heroLineA}</div>
- <div style={{ display: "flex", alignItems: "center" }}>
+ <div style={{ display: "flex", alignItems: "center", flexWrap: "wrap" }}>
<span>{SITE_CONFIG.heroLineB.prefix}</span>
<span
style={{
@@ -167,7 +168,7 @@ export default function OgImage(): ImageResponse {
padding: "1px 11px 3px",
color: palette.textSoft,
fontFamily: "monospace",
- fontSize: "30px",
+ fontSize: "34px",
lineHeight: 1.2,
}}
>
@@ -186,6 +187,34 @@ export default function OgImage(): ImageResponse {
<span>{SITE_CONFIG.heroLineB.period}</span>
</div>
</div>
+
+ <div
+ style={{
+ display: "flex",
+ gap: "10px",
+ marginTop: "26px",
+ }}
+ >
+ {SITE_CONFIG.harnessPillars.map((pillar) => (
+ <div
+ key={pillar}
+ style={{
+ display: "flex",
+ alignItems: "center",
+ borderRadius: "999px",
+ border: `1px solid ${palette.border}`,
+ backgroundColor: "rgba(255, 255, 255, 0.07)",
+ padding: "9px 14px",
+ color: palette.textSoft,
+ fontFamily: "monospace",
+ fontSize: "18px",
+ lineHeight: 1,
+ }}
+ >
+ {pillar}
+ </div>
+ ))}
+ </div>
</div>
</div>
diff --git a/packages/web/app/twitter-image.tsx b/packages/web/app/twitter-image.tsx
index abe77016..86a3d8cc 100644
--- a/packages/web/app/twitter-image.tsx
+++ b/packages/web/app/twitter-image.tsx
@@ -7,6 +7,6 @@
export { default } from "./opengraph-image"
export const runtime = "nodejs"
-export const alt = "LazyCodex - Codex for no-brainers."
+export const alt = "LazyCodex - agent harness for complex codebases."
export const size = { width: 1200, height: 630 }
export const contentType = "image/png"
diff --git a/packages/web/components/site/docs-cta.tsx b/packages/web/components/site/docs-cta.tsx
index 56a0d88a..75adcea3 100644
--- a/packages/web/components/site/docs-cta.tsx
+++ b/packages/web/components/site/docs-cta.tsx
@@ -5,18 +5,16 @@ import { SITE_CONFIG } from "../../lib/site-config"
export function DocsCta(): JSX.Element {
return (
<section className="mx-auto mt-32 mb-24 flex w-full max-w-[800px] flex-col items-center rounded-2xl border border-white/10 bg-[color:var(--surface-panel)] px-6 py-16 text-center md:mt-40 md:px-12 relative overflow-hidden">
- {/* Decorative cyan ring */}
<div className="pointer-events-none absolute left-1/2 top-1/2 -translate-x-1/2 -translate-y-1/2 h-[600px] w-[600px] rounded-full border-[1.5px] border-[rgba(135,240,242,0.12)]" />
-
+
<h2 className="relative z-10 text-3xl font-medium tracking-tight text-[color:var(--text-primary)]">
- Ready to stop thinking?
+ Ready to wire the harness?
</h2>
<p className="relative z-10 mt-4 text-lg text-[color:var(--text-muted)]">
- Read the manual. It's short.
+ Install the Codex setup, then run the workflows that keep large changes under control.
</p>
<div className="relative z-10 mt-8 flex flex-wrap justify-center gap-4">
<div className="relative group">
- {/* Soft cyan glow behind primary button */}
<div className="absolute -inset-1 rounded-lg bg-[color:var(--accent-glow)] opacity-20 blur-xl transition-opacity group-hover:opacity-30" />
<Link
href={SITE_CONFIG.docsPath}
diff --git a/packages/web/content/docs/installation.md b/packages/web/content/docs/installation.md
index e8e1b07c..f1c8890a 100644
--- a/packages/web/content/docs/installation.md
+++ b/packages/web/content/docs/installation.md
@@ -1,4 +1,4 @@
-One command gets you running. LazyCodex installs the Codex platform setup for OmO without a global package install.
+One command installs the OmO agent harness for Codex without a global package install.
### Install
diff --git a/packages/web/content/docs/overview.md b/packages/web/content/docs/overview.md
index ffe2015e..faf4587a 100644
--- a/packages/web/content/docs/overview.md
+++ b/packages/web/content/docs/overview.md
@@ -1,12 +1,12 @@
-LazyCodex is the lazy way to get [oh-my-openagent](https://github.com/code-yeongyu/oh-my-openagent) (OmO) up and running inside Codex. Think [LazyVim](https://github.com/LazyVim/LazyVim) for [lazy.nvim](https://github.com/folke/lazy.nvim), but for Codex.
+LazyCodex packages [oh-my-openagent](https://github.com/code-yeongyu/oh-my-openagent) (OmO) inside Codex as the agent harness for complex codebases. Think [LazyVim](https://github.com/LazyVim/LazyVim) for [lazy.nvim](https://github.com/folke/lazy.nvim), but for Codex.
### What you get
-OmO is a full AI dev team in a harness. It gives you discipline agents (Sisyphus orchestrates Hephaestus, Oracle, and Librarian), parallel execution, multi-model routing, a skills system, hooks and lifecycle, and zero-config defaults. LazyCodex wraps the entire setup so you don't have to think about it.
+OmO gives Codex a full agent harness: discipline agents (Sisyphus orchestrates Hephaestus, Oracle, and Librarian), parallel execution, multi-model routing, a skills system, hooks and lifecycle, and verification defaults. LazyCodex packages that harness as a repeatable Codex setup.
-### The one-word workflow
+### The harness workflow
-You don't need to ultrathink. Just prompt with `{your prompt} ultrawork`. One word, and every agent activates.
+Use `{your prompt} ultrawork` when the job needs project memory, planning, parallel agents, and verified completion to run as one coordinated loop.
### How it fits together
diff --git a/packages/web/content/docs/skills.md b/packages/web/content/docs/skills.md
index e80e8263..297e3545 100644
--- a/packages/web/content/docs/skills.md
+++ b/packages/web/content/docs/skills.md
@@ -1,4 +1,4 @@
-LazyCodex is most useful when you use the features it actually installs: project memory, planning, execution, verified completion, skills, hooks, model routing, and diagnostics.
+LazyCodex is most useful as a harness for complex codebases: project memory, planning, execution, verified completion, skills, hooks, model routing, and diagnostics.
### Built-in workflows
diff --git a/packages/web/content/docs/ultrawork.md b/packages/web/content/docs/ultrawork.md
index e7f22c25..1ef41be9 100644
--- a/packages/web/content/docs/ultrawork.md
+++ b/packages/web/content/docs/ultrawork.md
@@ -1,6 +1,6 @@
ultrawork is the headline mode. Typing `ultrawork` (or the short alias `ulw`) anywhere in your prompt activates maximum-precision, outcome-first, evidence-driven orchestration.
-> "One word. Every agent activates. Doesn't stop until done."
+> "Plan, execute, verify, and keep the evidence attached."
### Usage
diff --git a/packages/web/e2e/home.spec.ts b/packages/web/e2e/home.spec.ts
index df8777b2..0a6a01ce 100644
--- a/packages/web/e2e/home.spec.ts
+++ b/packages/web/e2e/home.spec.ts
@@ -23,7 +23,6 @@ test.describe("home page β€” content", () => {
await expect(page.getByText(SITE_CONFIG.eyebrow, { exact: true })).toBeVisible()
- // New hero copy (replaces the old "You don't need to think." splash).
await expect(page.getByText(SITE_CONFIG.heroLineA, { exact: false }).first()).toBeVisible()
await expect(
page.getByText(SITE_CONFIG.heroLineB.slot, { exact: false }).first(),
diff --git a/packages/web/e2e/landing.spec.ts b/packages/web/e2e/landing.spec.ts
index 78a37d12..b53dd1c9 100644
--- a/packages/web/e2e/landing.spec.ts
+++ b/packages/web/e2e/landing.spec.ts
@@ -7,7 +7,7 @@ import { SITE_CONFIG } from "../lib/site-config"
*
* Selectors are deliberately tolerant of inline <span> splits: headings,
* links, and buttons go through getByRole; the hero tagline is checked by
- * its distinctive sub-parts ("{your prompt}" + "ultrawork") instead of one
+ * its distinctive sub-parts ("planning" + "verified completion") instead of one
* exact full-string match; `.first()` guards against strict-mode violations.
*/
@@ -25,7 +25,9 @@ test.describe("landing page β€” hero", () => {
await expect(page.getByText(SITE_CONFIG.eyebrow, { exact: true })).toBeVisible()
await expect(page.getByText(SITE_CONFIG.heroLineA, { exact: false }).first()).toBeVisible()
// Tagline may be split across inline spans β€” assert the distinctive parts.
- await expect(page.getByText("Just prompt", { exact: false }).first()).toBeVisible()
+ await expect(
+ page.getByText(SITE_CONFIG.heroLineB.prefix, { exact: false }).first(),
+ ).toBeVisible()
await expect(
page.getByText(SITE_CONFIG.heroLineB.slot, { exact: false }).first(),
).toBeVisible()
@@ -58,10 +60,10 @@ test.describe("landing page β€” install + commands", () => {
await page.goto("/")
await expect(page.locator("article h2")).toHaveText(COMMANDS.map((command) => command.name))
- await expect(page.getByText("Use the built-in workflows", { exact: false })).toBeVisible()
- await expect(page.getByText("Project memory", { exact: false }).first()).toBeVisible()
- await expect(page.getByText("Planning and execution", { exact: false }).first()).toBeVisible()
- await expect(page.getByText("Verified completion", { exact: false }).first()).toBeVisible()
+ await expect(page.getByText("Harness the whole codebase", { exact: false })).toBeVisible()
+ await expect(page.getByText("Context that survives", { exact: false }).first()).toBeVisible()
+ await expect(page.getByText("Plans before edits", { exact: false }).first()).toBeVisible()
+ await expect(page.getByText("Evidence at the end", { exact: false }).first()).toBeVisible()
await expect(page.getByText("Built-in skill coverage", { exact: false }).first()).toBeVisible()
})
})
diff --git a/packages/web/e2e/responsive.spec.ts b/packages/web/e2e/responsive.spec.ts
index 6c819bb4..507bc6b6 100644
--- a/packages/web/e2e/responsive.spec.ts
+++ b/packages/web/e2e/responsive.spec.ts
@@ -9,7 +9,7 @@ import { COMMANDS } from "../lib/commands"
* For each viewport, we assert:
* 1. No horizontal overflow (the page fits).
* 2. The wordmark is visible AND fits inside the viewport width.
- * 3. The hero keyword "ultrawork" is visible.
+ * 3. The hero keyword is visible.
* 4. All three command names are visible.
* 5. The footer ("lazycodex.ai") is visible.
*
diff --git a/packages/web/e2e/seo.spec.ts b/packages/web/e2e/seo.spec.ts
index bddf0376..750d171a 100644
--- a/packages/web/e2e/seo.spec.ts
+++ b/packages/web/e2e/seo.spec.ts
@@ -32,6 +32,8 @@ test.describe("site SEO + metadata", () => {
expect(description?.length).toBeGreaterThan(50)
expect(description?.length).toBeLessThanOrEqual(170)
expect(description).toMatch(/Codex/i)
+ expect(description).toMatch(/agent harness/i)
+ expect(description).toMatch(/complex codebases/i)
expect(description).not.toMatch(LAUNCH_GATING_PATTERN)
const canonical = await page.locator('link[rel="canonical"]').getAttribute("href")
diff --git a/packages/web/lib/commands.ts b/packages/web/lib/commands.ts
index 88f65cd4..44cc5132 100644
--- a/packages/web/lib/commands.ts
+++ b/packages/web/lib/commands.ts
@@ -13,7 +13,7 @@ export const COMMANDS: readonly LazyCommand[] = [
syntax: '/ulw-loop "task" [--completion-promise=TEXT] [--strategy=reset|continue]',
summary: "A self-referential loop that runs until VERIFIED completion.",
facts: [
- "Emits <promise>DONE</promise> when it thinks it's done",
+ "Emits <promise>DONE</promise> when it reaches a completion claim",
"Oracle must verify before the loop ends",
"Iteration cap 500 in ultrawork mode (100 normal)",
],
diff --git a/packages/web/lib/docs-content.generated.ts b/packages/web/lib/docs-content.generated.ts
index 3e1b335d..6ef40e3f 100644
--- a/packages/web/lib/docs-content.generated.ts
+++ b/packages/web/lib/docs-content.generated.ts
@@ -1,9 +1,9 @@
// AUTO-GENERATED β€” do not edit. Run: node ./scripts/generate-docs-content.mjs
export const DOC_SOURCES: Record<string, string> = {
- "overview.md": "<p>LazyCodex is the lazy way to get <a href=\"https://github.com/code-yeongyu/oh-my-openagent\">oh-my-openagent</a> (OmO) up and running inside Codex. Think <a href=\"https://github.com/LazyVim/LazyVim\">LazyVim</a> for <a href=\"https://github.com/folke/lazy.nvim\">lazy.nvim</a>, but for Codex.</p>\n<h3>What you get</h3>\n<p>OmO is a full AI dev team in a harness. It gives you discipline agents (Sisyphus orchestrates Hephaestus, Oracle, and Librarian), parallel execution, multi-model routing, a skills system, hooks and lifecycle, and zero-config defaults. LazyCodex wraps the entire setup so you don&#39;t have to think about it.</p>\n<h3>The one-word workflow</h3>\n<p>You don&#39;t need to ultrathink. Just prompt with <code>{your prompt} ultrawork</code>. One word, and every agent activates.</p>\n<h3>How it fits together</h3>\n<p>LazyCodex is a thin distribution layer. The core engine is <a href=\"https://github.com/code-yeongyu/oh-my-openagent\">OmO</a>. LazyCodex is maintained by <a href=\"https://sisyphuslabs.ai\">Sisyphus Labs</a>.</p>\n<ul>\n<li><a href=\"https://github.com/code-yeongyu/lazycodex\">LazyCodex on GitHub</a></li>\n<li><a href=\"https://github.com/code-yeongyu/oh-my-openagent\">OmO on GitHub</a></li>\n</ul>\n",
- "installation.md": "<p>One command gets you running. LazyCodex installs the Codex platform setup for OmO without a global package install.</p>\n<h3>Install</h3>\n<pre><code class=\"language-bash\">npx lazycodex-ai install\n</code></pre>\n<p>This is exactly equivalent to <code>npx --yes --package oh-my-openagent omo install --platform=codex</code>.</p>\n<h3>Autonomous one-liner</h3>\n<pre><code class=\"language-bash\">npx lazycodex-ai install --no-tui --codex-autonomous\n</code></pre>\n<h3>Prerequisites</h3>\n<ul>\n<li><a href=\"https://bun.sh\">Bun</a></li>\n<li>The <a href=\"https://github.com/openai/codex\">OpenAI Codex CLI</a></li>\n</ul>\n<blockquote>\n<p>Do NOT use <code>npm install -g</code> or <code>bun add -g</code>. Always invoke via <code>npx</code>.</p>\n</blockquote>\n<h3>Let an agent do it</h3>\n<p>It is strongly recommended to let an LLM agent run the install and walk the setup for you. The agent handles subscription detection, model selection, and provider auth automatically.</p>\n",
- "skills.md": "<p>LazyCodex is most useful when you use the features it actually installs: project memory, planning, execution, verified completion, skills, hooks, model routing, and diagnostics.</p>\n<h3>Built-in workflows</h3>\n<p>Start with <code>/init-deep</code> when the repository is too large or too old to explain from memory. It generates hierarchical <code>AGENTS.md</code> context so agents can find the right files before they change code.</p>\n<p>Use <code>$ulw-plan</code> when the work needs decisions before implementation, <code>$start-work</code> when a plan should be executed, and <code>$ulw-loop</code> when you want the agent to keep going until the result is verified.</p>\n<h3>Feature coverage</h3>\n<p>The three command pillars stay simple:</p>\n<ul>\n<li><code>$ulw-loop</code> keeps moving until verified completion</li>\n<li><code>$ulw-plan</code> turns fuzzy work into a decision-complete plan</li>\n<li><code>$start-work</code> executes a plan with durable Boulder progress</li>\n</ul>\n<p>Skills add specialist judgment around those pillars:</p>\n<table>\n<thead>\n<tr>\n<th>Skill</th>\n<th>Use it for</th>\n</tr>\n</thead>\n<tbody><tr>\n<td><code>review-work</code></td>\n<td>Multi-angle post-implementation review</td>\n</tr>\n<tr>\n<td><code>remove-ai-slops</code></td>\n<td>Behavior-preserving cleanup of AI-looking code</td>\n</tr>\n<tr>\n<td><code>frontend-ui-ux</code></td>\n<td>Designed UI work instead of generic layout filling</td>\n</tr>\n<tr>\n<td><code>programming</code></td>\n<td>Strict TypeScript, Rust, Python, or Go discipline</td>\n</tr>\n<tr>\n<td><code>LSP</code></td>\n<td>Diagnostics, definitions, references, symbols, and renames</td>\n</tr>\n<tr>\n<td><code>AST-grep</code></td>\n<td>Structural search and rewrite across code</td>\n</tr>\n<tr>\n<td><code>rules</code></td>\n<td>Project instructions from AGENTS, rules, and instruction files</td>\n</tr>\n<tr>\n<td><code>comment-checker</code></td>\n<td>Feedback after edit-like operations</td>\n</tr>\n</tbody></table>\n<h3>Where skills live</h3>\n<p>OmO can load skills from project and user locations such as <code>.opencode/skills</code>, <code>~/.config/opencode/skills</code>, <code>.claude/skills</code>, <code>.agents/skills</code>, and <code>~/.agents/skills</code>.</p>\n<p>LazyCodex installs the Codex Light setup with:</p>\n<pre><code class=\"language-bash\">npx lazycodex-ai install\n</code></pre>\n<p>That installer wires the Codex marketplace plugin as <code>omo@sisyphuslabs</code> while keeping the public package alias easy to remember.</p>\n",
- "ultrawork.md": "<p>ultrawork is the headline mode. Typing <code>ultrawork</code> (or the short alias <code>ulw</code>) anywhere in your prompt activates maximum-precision, outcome-first, evidence-driven orchestration.</p>\n<blockquote>\n<p>&quot;One word. Every agent activates. Doesn&#39;t stop until done.&quot;</p>\n</blockquote>\n<h3>Usage</h3>\n<pre><code class=\"language-bash\">ulw add authentication\n</code></pre>\n<h3>What it enforces</h3>\n<ul>\n<li>Strict TDD: RED β†’ GREEN β†’ SURFACE β†’ CLEAN</li>\n<li>At least 3 realistic QA scenarios</li>\n<li>Real manual-QA channels (HTTP call, tmux, browser)</li>\n<li>A binding verification gate that loops until the work is genuinely done</li>\n</ul>\n",
+ "overview.md": "<p>LazyCodex packages <a href=\"https://github.com/code-yeongyu/oh-my-openagent\">oh-my-openagent</a> (OmO) inside Codex as the agent harness for complex codebases. Think <a href=\"https://github.com/LazyVim/LazyVim\">LazyVim</a> for <a href=\"https://github.com/folke/lazy.nvim\">lazy.nvim</a>, but for Codex.</p>\n<h3>What you get</h3>\n<p>OmO gives Codex a full agent harness: discipline agents (Sisyphus orchestrates Hephaestus, Oracle, and Librarian), parallel execution, multi-model routing, a skills system, hooks and lifecycle, and verification defaults. LazyCodex packages that harness as a repeatable Codex setup.</p>\n<h3>The harness workflow</h3>\n<p>Use <code>{your prompt} ultrawork</code> when the job needs project memory, planning, parallel agents, and verified completion to run as one coordinated loop.</p>\n<h3>How it fits together</h3>\n<p>LazyCodex is a thin distribution layer. The core engine is <a href=\"https://github.com/code-yeongyu/oh-my-openagent\">OmO</a>. LazyCodex is maintained by <a href=\"https://sisyphuslabs.ai\">Sisyphus Labs</a>.</p>\n<ul>\n<li><a href=\"https://github.com/code-yeongyu/lazycodex\">LazyCodex on GitHub</a></li>\n<li><a href=\"https://github.com/code-yeongyu/oh-my-openagent\">OmO on GitHub</a></li>\n</ul>\n",
+ "installation.md": "<p>One command installs the OmO agent harness for Codex without a global package install.</p>\n<h3>Install</h3>\n<pre><code class=\"language-bash\">npx lazycodex-ai install\n</code></pre>\n<p>This is exactly equivalent to <code>npx --yes --package oh-my-openagent omo install --platform=codex</code>.</p>\n<h3>Autonomous one-liner</h3>\n<pre><code class=\"language-bash\">npx lazycodex-ai install --no-tui --codex-autonomous\n</code></pre>\n<h3>Prerequisites</h3>\n<ul>\n<li><a href=\"https://bun.sh\">Bun</a></li>\n<li>The <a href=\"https://github.com/openai/codex\">OpenAI Codex CLI</a></li>\n</ul>\n<blockquote>\n<p>Do NOT use <code>npm install -g</code> or <code>bun add -g</code>. Always invoke via <code>npx</code>.</p>\n</blockquote>\n<h3>Let an agent do it</h3>\n<p>It is strongly recommended to let an LLM agent run the install and walk the setup for you. The agent handles subscription detection, model selection, and provider auth automatically.</p>\n",
+ "skills.md": "<p>LazyCodex is most useful as a harness for complex codebases: project memory, planning, execution, verified completion, skills, hooks, model routing, and diagnostics.</p>\n<h3>Built-in workflows</h3>\n<p>Start with <code>/init-deep</code> when the repository is too large or too old to explain from memory. It generates hierarchical <code>AGENTS.md</code> context so agents can find the right files before they change code.</p>\n<p>Use <code>$ulw-plan</code> when the work needs decisions before implementation, <code>$start-work</code> when a plan should be executed, and <code>$ulw-loop</code> when you want the agent to keep going until the result is verified.</p>\n<h3>Feature coverage</h3>\n<p>The three command pillars stay simple:</p>\n<ul>\n<li><code>$ulw-loop</code> keeps moving until verified completion</li>\n<li><code>$ulw-plan</code> turns fuzzy work into a decision-complete plan</li>\n<li><code>$start-work</code> executes a plan with durable Boulder progress</li>\n</ul>\n<p>Skills add specialist judgment around those pillars:</p>\n<table>\n<thead>\n<tr>\n<th>Skill</th>\n<th>Use it for</th>\n</tr>\n</thead>\n<tbody><tr>\n<td><code>review-work</code></td>\n<td>Multi-angle post-implementation review</td>\n</tr>\n<tr>\n<td><code>remove-ai-slops</code></td>\n<td>Behavior-preserving cleanup of AI-looking code</td>\n</tr>\n<tr>\n<td><code>frontend-ui-ux</code></td>\n<td>Designed UI work instead of generic layout filling</td>\n</tr>\n<tr>\n<td><code>programming</code></td>\n<td>Strict TypeScript, Rust, Python, or Go discipline</td>\n</tr>\n<tr>\n<td><code>LSP</code></td>\n<td>Diagnostics, definitions, references, symbols, and renames</td>\n</tr>\n<tr>\n<td><code>AST-grep</code></td>\n<td>Structural search and rewrite across code</td>\n</tr>\n<tr>\n<td><code>rules</code></td>\n<td>Project instructions from AGENTS, rules, and instruction files</td>\n</tr>\n<tr>\n<td><code>comment-checker</code></td>\n<td>Feedback after edit-like operations</td>\n</tr>\n</tbody></table>\n<h3>Where skills live</h3>\n<p>OmO can load skills from project and user locations such as <code>.opencode/skills</code>, <code>~/.config/opencode/skills</code>, <code>.claude/skills</code>, <code>.agents/skills</code>, and <code>~/.agents/skills</code>.</p>\n<p>LazyCodex installs the Codex Light setup with:</p>\n<pre><code class=\"language-bash\">npx lazycodex-ai install\n</code></pre>\n<p>That installer wires the Codex marketplace plugin as <code>omo@sisyphuslabs</code> while keeping the public package alias easy to remember.</p>\n",
+ "ultrawork.md": "<p>ultrawork is the headline mode. Typing <code>ultrawork</code> (or the short alias <code>ulw</code>) anywhere in your prompt activates maximum-precision, outcome-first, evidence-driven orchestration.</p>\n<blockquote>\n<p>&quot;Plan, execute, verify, and keep the evidence attached.&quot;</p>\n</blockquote>\n<h3>Usage</h3>\n<pre><code class=\"language-bash\">ulw add authentication\n</code></pre>\n<h3>What it enforces</h3>\n<ul>\n<li>Strict TDD: RED β†’ GREEN β†’ SURFACE β†’ CLEAN</li>\n<li>At least 3 realistic QA scenarios</li>\n<li>Real manual-QA channels (HTTP call, tmux, browser)</li>\n<li>A binding verification gate that loops until the work is genuinely done</li>\n</ul>\n",
"ulw-loop.md": "<p><code>$ulw-loop</code> is a self-referential development loop that runs until verified completion.</p>\n<h3>How it works</h3>\n<p>The agent works continuously and emits <code>&lt;promise&gt;DONE&lt;/promise&gt;</code> when it believes the task is complete, but that does NOT end the loop. An Oracle must verify the result first. The loop ends only after the system confirms Oracle verified it. If verification fails, it continues with the message: &quot;Oracle verification failed. Continuing ULTRAWORK loop.&quot;</p>\n<h3>Syntax</h3>\n<pre><code class=\"language-bash\">/ulw-loop &quot;task description&quot; [--completion-promise=TEXT] [--strategy=reset|continue]\n</code></pre>\n<h3>Limits</h3>\n<p>The iteration cap is 500 in ultrawork mode (100 in normal mode).</p>\n",
"ulw-plan.md": "<p><code>$ulw-plan</code> is the strategic planning consultant (Prometheus). It turns an idea into a decision-complete work plan. It is a planner, NOT an implementer. When you say &quot;do X&quot; it produces a plan for X and never writes product code.</p>\n<h3>The flow</h3>\n<ol>\n<li>Socratic interview</li>\n<li>Parallel codebase exploration</li>\n<li>Metis gap analysis</li>\n<li>Writes the plan to <code>plans/&lt;slug&gt;.md</code></li>\n<li>Optional Momus high-accuracy review</li>\n</ol>\n<h3>Output</h3>\n<p>Questions, research, and a work plan you can hand to <a href=\"#start-work\"><code>$start-work</code></a>.</p>\n",
"start-work.md": "<p><code>$start-work</code> executes a Prometheus work plan until every top-level checkbox is done.</p>\n<h3>How it works</h3>\n<ul>\n<li>Durable Boulder state in <code>.omo/boulder.json</code> survives across turns and sessions</li>\n<li>A Stop-hook re-injects the next turn until the plan is complete</li>\n<li>Independent sub-tasks fan out to parallel subagents</li>\n<li>Strict TDD plus five evidence gates: plan reread, automated verification, manual-QA, adversarial QA, cleanup</li>\n<li>Progress is recorded to a ledger</li>\n</ul>\n<h3>Syntax</h3>\n<pre><code class=\"language-bash\">/start-work [plan-name] [--worktree &lt;absolute-path&gt;]\n</code></pre>\n<h3>Done</h3>\n<p>It prints an <code>ORCHESTRATION COMPLETE</code> block when every checkbox is checked.</p>\n"
diff --git a/packages/web/lib/site-config.ts b/packages/web/lib/site-config.ts
index 8cc66e34..1e52901e 100644
--- a/packages/web/lib/site-config.ts
+++ b/packages/web/lib/site-config.ts
@@ -8,42 +8,43 @@ export const SITE_CONFIG = {
sisyphusUrl: "https://sisyphuslabs.ai",
siteUrl: "https://lazycodex.ai",
docsPath: "/docs",
- eyebrow: "CODEX FOR NO-BRAINERS",
+ eyebrow: "AGENT HARNESS FOR COMPLEX CODEBASES",
wordmark: "LazyCodex",
- heroLineA: "You don't need to ultrathink.",
+ heroLineA: "The one and only agent harness for complex codebases.",
heroLineB: {
- prefix: "Just prompt ",
- slot: "{your prompt}",
- suffix: " ",
- keyword: "ultrawork",
+ prefix: "Project memory, ",
+ slot: "planning,",
+ suffix: " execution, and ",
+ keyword: "verified completion",
period: ".",
},
- ultraworkTagline: "One word. Every agent activates. Doesn't stop until done.",
+ harnessPillars: ["project memory", "planning", "parallel agents", "verified completion"],
+ ultraworkTagline: "ultrawork turns the harness into a verified run.",
ultraworkExample: "ulw add authentication",
featureWorkflows: {
- kicker: "What LazyCodex installs",
- title: "Use the built-in workflows",
+ kicker: "What LazyCodex wires into Codex",
+ title: "Harness the whole codebase",
intro:
- "LazyCodex is a thin Codex distribution for OmO. It installs the command loop, planner, executor, project-memory setup, skills, hooks, model routing, and verification defaults in one pass.",
+ "LazyCodex installs OmO as a serious agent harness for complex repositories: project memory, planning, execution, skills, hooks, model routing, and verification defaults in one pass.",
points: [
{
- label: "Project memory",
- text: "/init-deep generates hierarchical AGENTS.md context so future agents start with local guidance instead of guessing.",
+ label: "Context that survives",
+ text: "/init-deep generates hierarchical AGENTS.md context so agents start from local guidance before touching a large repository.",
},
{
- label: "Planning and execution",
- text: "$ulw-plan turns fuzzy work into a plan, then $start-work executes that plan with durable Boulder progress.",
+ label: "Plans before edits",
+ text: "$ulw-plan turns ambiguous work into a decision-complete plan, then $start-work executes it with durable Boulder progress.",
},
{
- label: "Verified completion",
- text: "$ulw-loop keeps work moving until the completion promise is backed by evidence instead of a hopeful status update.",
+ label: "Evidence at the end",
+ text: "$ulw-loop keeps complex work moving until the completion promise is backed by verification, not a hopeful status update.",
},
],
},
builtInSkills: {
title: "Built-in skill coverage",
summary:
- "Skills add specialist guidance for review, cleanup, UI, strict language work, LSP diagnostics, AST-grep rewrites, project rules, and comment-checker feedback.",
+ "Skills give the harness specialist judgment for review, cleanup, UI, strict language work, LSP diagnostics, AST-grep rewrites, project rules, and comment-checker feedback.",
skills: ["review-work", "remove-ai-slops", "frontend-ui-ux", "programming", "LSP", "AST-grep", "rules", "comment-checker"],
},
} as const;