Skip to content

Commit 585cc06

Browse files
authored
Merge pull request #10 from secondorderai/dev
Dev
2 parents 0bcf1b2 + 1985f68 commit 585cc06

16 files changed

Lines changed: 956 additions & 237 deletions

AGENTS.md

Lines changed: 21 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,21 @@
1+
# AGENTS.md
2+
3+
SecondOrder is a Next.js App Router web app for exploring meta-thinking workflows, with both a marketing site and a `/chat` AI assistant.
4+
5+
## Essentials
6+
7+
- Node.js: `24.12` (see [`.nvmrc`](/Users/henry/workspace/secondorder-web/.nvmrc))
8+
- Typecheck: `npm run ts-check`
9+
- The app uses the Next.js App Router.
10+
- TypeScript strict mode is enabled.
11+
- Use the `@/*` import alias for internal imports.
12+
13+
## Progressive Disclosure
14+
15+
Load only the file that matches the task instead of reading all project guidance up front:
16+
17+
- [Architecture](/Users/henry/workspace/secondorder-web/docs/agents/architecture.md)
18+
- [Styling](/Users/henry/workspace/secondorder-web/docs/agents/styling.md)
19+
- [Testing](/Users/henry/workspace/secondorder-web/docs/agents/testing.md)
20+
- [UI Patterns](/Users/henry/workspace/secondorder-web/docs/agents/ui-patterns.md)
21+
- [Product and Content](/Users/henry/workspace/secondorder-web/docs/agents/product-and-content.md)

CLAUDE.md

Lines changed: 0 additions & 114 deletions
This file was deleted.

README.md

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -2,7 +2,7 @@
22

33
# Features
44

5-
- Integrate Vercel AI chat AI with AI gateway gpt-5-mini, gpt-5.2, gpt-5.1-thinking, with OpenAI API Key $19 credits
5+
- Integrate Vercel AI chat AI with AI gateway gpt-5.4
66
- Implement basic meta thinking prompt, dynamic thinking layer
77
- Solver builder and Solver with Tools
88
- Create a multi-step, self-improving process, making a system that autonomously audits its own progress

app/api/chat/route.ts

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -78,7 +78,7 @@ export async function POST(req: Request) {
7878
}
7979

8080
const result = streamText({
81-
model: openai('gpt-5.2'),
81+
model: openai('gpt-5.4'),
8282
system: SYSTEM_PROMPT,
8383
messages: normalizedMessages,
8484
});
@@ -93,7 +93,7 @@ export async function POST(req: Request) {
9393
}
9494
return new Response(
9595
JSON.stringify({ error: 'An unexpected error occurred' }),
96-
{ status: 500, headers: { 'Content-Type': 'application/json' } }
96+
{ status: 500, headers: { 'Content-Type': 'application/json' } },
9797
);
9898
}
9999
}

app/page.tsx

Lines changed: 111 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -46,6 +46,48 @@ const overview = [
4646
'Context engineering automated end to end.',
4747
];
4848

49+
const implementedToday = [
50+
{
51+
label: 'Chat experience',
52+
title: 'Streaming assistant in /chat',
53+
description:
54+
'A working chat interface already ships with streaming responses and a focused meta-thinking assistant persona.',
55+
},
56+
{
57+
label: 'Core model',
58+
title: 'GPT-5.4-backed reasoning',
59+
description:
60+
'The current assistant runs on GPT-5.4 with a system prompt centered on planning, self-evaluation, and adaptive reasoning.',
61+
},
62+
{
63+
label: 'Interaction loop',
64+
title: 'Prompt-to-response workflow',
65+
description:
66+
'Users can already move from the marketing page into a live assistant experience designed around reflective problem solving.',
67+
},
68+
];
69+
70+
const roadmapPhases = [
71+
{
72+
phase: 'Phase 1',
73+
title: 'Structured response modes',
74+
description:
75+
'Introduce explicit plan, answer, critique, and refine behaviors so the assistant acts more deliberately on every task.',
76+
},
77+
{
78+
phase: 'Phase 2',
79+
title: 'Planning and self-critique',
80+
description:
81+
'Add planner and critic passes that extract constraints, inspect drafts, and improve answer quality before delivery.',
82+
},
83+
{
84+
phase: 'Phase 3',
85+
title: 'Tools, memory, and orchestration',
86+
description:
87+
'Expand beyond chat with tool routing, memory, and eventually a judge-agent layer that chooses stronger strategies per task.',
88+
},
89+
];
90+
4991
export default function Home() {
5092
return (
5193
<main className="min-h-screen bg-bone text-ink">
@@ -106,7 +148,7 @@ export default function Home() {
106148
<p className="text-sm uppercase tracking-[0.3em] text-ink/50">
107149
AI gateway
108150
</p>
109-
<p className="mt-2 text-xl font-semibold">GPT-5.2 stack</p>
151+
<p className="mt-2 text-xl font-semibold">GPT-5.4 stack</p>
110152
</div>
111153
<div>
112154
<p className="text-sm uppercase tracking-[0.3em] text-ink/50">
@@ -177,6 +219,38 @@ export default function Home() {
177219
</div>
178220
</section>
179221

222+
<section className="mx-auto max-w-6xl px-6 py-16 sm:px-10">
223+
<div className="grid gap-10 lg:grid-cols-[0.95fr_1.05fr]">
224+
<div>
225+
<p className="text-sm uppercase tracking-[0.4em] text-ink/60">
226+
Implemented today
227+
</p>
228+
<h2 className="mt-4 font-display text-4xl">
229+
A working v1 is already in place.
230+
</h2>
231+
<p className="mt-4 text-ink/70">
232+
SecondOrder is not just a concept page. The current release
233+
already includes a live chat assistant, streaming responses, and a
234+
meta-thinking system prompt focused on reflective reasoning.
235+
</p>
236+
</div>
237+
<div className="grid gap-4">
238+
{implementedToday.map((item) => (
239+
<div
240+
key={item.title}
241+
className="rounded-3xl border border-ink/10 bg-white/70 p-6"
242+
>
243+
<p className="text-xs uppercase tracking-[0.3em] text-ink/55">
244+
{item.label}
245+
</p>
246+
<h3 className="mt-3 text-xl font-semibold">{item.title}</h3>
247+
<p className="mt-3 text-sm text-ink/75">{item.description}</p>
248+
</div>
249+
))}
250+
</div>
251+
</div>
252+
</section>
253+
180254
<section className="mx-auto max-w-6xl px-6 py-16 sm:px-10">
181255
<div className="grid gap-10 lg:grid-cols-2">
182256
<div className="rounded-3xl border border-ink/10 bg-white/70 p-8">
@@ -259,6 +333,42 @@ export default function Home() {
259333
</div>
260334
</section>
261335

336+
<section className="mx-auto max-w-6xl px-6 py-16 sm:px-10">
337+
<div className="rounded-3xl border border-ink/10 bg-ink p-8 text-bone sm:p-12">
338+
<div className="grid gap-8 lg:grid-cols-[0.9fr_1.1fr]">
339+
<div>
340+
<p className="text-sm uppercase tracking-[0.4em] text-bone/70">
341+
Product roadmap
342+
</p>
343+
<h2 className="mt-4 font-display text-4xl">
344+
What comes after the current assistant.
345+
</h2>
346+
<p className="mt-4 text-bone/75">
347+
The next releases push SecondOrder from a prompt-shaped
348+
assistant toward a real meta-cognitive system with planning,
349+
critique, tools, memory, and orchestration.
350+
</p>
351+
</div>
352+
<div className="space-y-4">
353+
{roadmapPhases.map((item) => (
354+
<div
355+
key={item.title}
356+
className="rounded-2xl border border-bone/15 bg-bone/5 p-5"
357+
>
358+
<p className="text-xs uppercase tracking-[0.3em] text-bone/55">
359+
{item.phase}
360+
</p>
361+
<h3 className="mt-2 text-xl font-semibold">{item.title}</h3>
362+
<p className="mt-3 text-sm text-bone/75">
363+
{item.description}
364+
</p>
365+
</div>
366+
))}
367+
</div>
368+
</div>
369+
</div>
370+
</section>
371+
262372
<section className="border-t border-ink/10 bg-bone" id="ready-to-build">
263373
<div className="mx-auto max-w-6xl px-6 py-16 sm:px-10">
264374
<div className="flex flex-col items-start justify-between gap-6 md:flex-row md:items-center">

0 commit comments

Comments
 (0)