aichitect: Instructions file for Claude Code

CLAUDE.md

aichitect CLAUDE.md is an instructions file for Claude Code from letsgogeeky/aichitect. It costs 6,521 tokens per session, scanned A, original, MIT.

Project instructions for AIchitect, a catalogue application for comparing AI tools and services. They describe its technology stack, engineering rules, development process, and the file that defines which tools belong in the catalogue.

In plain words
What is it for?
Checking catalogue scope, following the project's clean-code conventions, working with its web interface and data flow, and using the prescribed development and review process.
Why use it?
They help the coding agent make changes that fit the existing application and avoid adding tools or compatibility code outside the project's intended scope.

Instructions file for Claude Code

Written for Claude Code: the file is CLAUDE.md. Also seen: mentions CLAUDE.md; mentions AGENTS.md; built for aider.

This is letsgogeeky/aichitect's own configuration. It tells Claude Code how to work on aichitect itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything aichitect configures →

Reuse

Borrowing it

Nothing to install: this file belongs to letsgogeeky/aichitect. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/letsgogeeky/aichitect/main/CLAUDE.md
Clone the repo
git clone --depth 1 https://github.com/letsgogeeky/aichitect

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for aichitect CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/letsgogeeky/aichitect/claude-md/github.svg)](https://agentmods.dev/instructions/letsgogeeky/aichitect/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/letsgogeeky/aichitect/claude-md"><img src="https://agentmods.dev/badge/instructions/letsgogeeky/aichitect/claude-md/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for aichitect CLAUDE.md

Your own site · 80×15
<a href="https://agentmods.dev/instructions/letsgogeeky/aichitect/claude-md"><img src="https://agentmods.dev/badge/instructions/letsgogeeky/aichitect/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 6,521 This file is loaded in full into every session.
When invoked 6,521 The same file — it is already loaded in full.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.06521 $0.06521
Opus 5.5 $0.02608 $0.02608
Sonnet 5.5 $0.01304 $0.01304
Haiku 4.5 $0.00652 $0.00652

Measured 29d ago against content hash c826940ba830, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

aichitect CLAUDE.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 29d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

CLAUDE.md · 424 lines

How it starts

The opening of the file, as written. The whole thing — 424 lines — stays where its author put it; the contents beside it link to each section on GitHub.

@AGENTS.md

AIchitect — Project Context

Brand: "Cut the noise. Pick your AI stack." Domain: aichitect.dev | Stack: Next.js 16 + React Flow + Three.js + Tailwind v4 + TypeScript Scope rule: SCOPE.md — the single source of truth for what belongs in the catalog. Read it before adding a tool.

Deployment

Live on Cloudflare Workers (aichitect-web) via OpenNext as of 2026-10-02 — tracked in Linear [AIC-182], same playbook proven on Thoth (2026-09-23) and Sopdet (2026-09-29). aichitect.dev, www.aichitect.dev, and aichitect-staging.aichitect.dev all serve the Worker. Vercel's deployment was already down (402 DEPLOYMENT_DISABLED) by the time of the cutover, so this fixed an outage rather than disrupting live traffic.

  • Deploy: make deploy → scripts/deploy.sh, run on the host, not Docker (OpenNext bakes absolute build paths into the bundle; workerd needs the host's native build). Manual by choice, like Thoth and Sopdet. next build inlines the NEXT_PUBLIC_POSTGRES_SUPABASE_* values from .env.local; runtime secrets are Worker secrets (wrangler secret put).
  • Analytics: @vercel/analytics / speed-insights removed (they only report through Vercel). PostHog stays; Cloudflare Web Analytics is enabled per hostname in the Cloudflare dashboard (automatic setup, no code).
  • Database: staying on Supabase entirely — verified empirically that its REST API (PostgREST) works fine from workerd, unlike Thoth/Sopdet which hit a raw-Postgres-TCP incompatibility and had to move to PlanetScale. This app never makes a direct Postgres connection at runtime, so that problem doesn't apply here. No database migration happened or is needed.
  • Crons: worker.ts provides a custom scheduled() handler (via a self-referencing SELF service binding) dispatching to the same /api/cron/* routes Vercel Cron used to hit — @opennextjs/cloudflare exports no scheduled() of its own, so wrangler.jsonc's triggers.crons would otherwise silently no-op. All 3 crons (sync-health, sync-benchmarks, sync-incidents) verified firing correctly post-cutover.
  • export const runtime = "edge" was removed from every OG/badge route — OpenNext-Cloudflare runs everything in a single Workers runtime; the Vercel edge/node split doesn't apply the same way. See "Edge runtime & SSR rules" below, which is being revised accordingly.
  • GITHUB_TOKEN deliberately deferred (Ramy's call, 2026-10-02) — not set as a Worker secret. sync-health runs unauthenticated (60 req/hr) and effectively processes nothing until this is set. Needs a GitHub PAT with read-only/public-repo scope.
  • Two pre-existing bugs found in sync-incidents, unrelated to this migration (would fail identically on Vercel): some Cohere incidents have a null shortlink, which violates tool_incidents.url's NOT NULL constraint; Replicate's configured status-page URL now returns HTML instead of JSON (stale URL, needs updating in whatever table configures status_pages).
  • GitHub OAuth continues working unchanged post-cutover — Supabase's Redirect URLs allowlist is keyed to the domain (aichitect.dev), not the hosting platform, so the existing entry from the Vercel era still applies. The staging subdomain is a separate, lower-priority gap (OAuth doesn't work there — https://aichitect-staging.aichitect.dev/** was never added to Supabase's allowlist — a dashboard change, not code).
  • terraform/ — own directory and state, not shared with Thoth's or Sopdet's.

Read the full file on GitHub · 424 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 29d ago Changed · +14 lines · +970 tokens per session c826940ba830
  2. 2mo ago First seen · 410 lines · 5,551 tokens per session scan A adb8fde74a54

Subscribe to this mod's changes

aichitect CLAUDE.md is an instructions file published in the GitHub repository letsgogeeky/aichitect (5 stars, last pushed 1mo ago), licensed MIT. It adds 6,521 tokens to every session, about $0.0261 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

aichitect CLAUDE.md compared with instructions from other repositories

aichitect CLAUDE.md next to 6 instructions from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
aichitect CLAUDE.mdletsgogeeky/aichitectInstructions file51mo agoMITA6,521Claude Code
vscode buildNext.instructions.mdmicrosoft/vscodeInstructions file195ktodayMITA9,719GitHub Copilot
next.js AGENTS.mdvercel/next.jsInstructions file143kyesterdayMITA7,461Codex, OpenCode
vscode oss-third-party-notices.instructions.mdmicrosoft/vscodeInstructions file195ktodayMITA5,649GitHub Copilot
langchain AGENTS.mdlangchain-ai/langchainInstructions file149ktodayMITA4,680Codex, OpenCode
vscode copilot-instructions.mdmicrosoft/vscodeInstructions file195ktodayMITA3,093GitHub Copilot
vscode ai-customization.instructions.mdmicrosoft/vscodeInstructions file195ktodayMITA1,907GitHub Copilot