vibecodingisold¶

The missing step between Context Engineering and AI Planning. 314 senior-engineering Mandatory Items injected before any AI tool plans your build.

AI-coding tools generate code well. They do not generate the right requirements. The same scaffold ships to production over and over: auto-increment integer primary keys, JWT in localStorage, no idempotency on payment endpoints, no tenant_id column ("we'll multi-tenant later"), no rate limit on /login, no structured logs, no rollback plan. Each defect is forgivable in isolation. Stacked, they're a six-week rebuild that the founder will do under time pressure when a paying customer reports a double-charge.

vibecodingisold is the layer that closes the gap. It sits between your context-engineering output (PRD, vision doc, stack constraints) and your planning tool of choice (Claude Code, Cursor, Cline, Aider, ChatGPT, plain Markdown). It hands the planning tool the 200+ items a senior engineering team would have demanded, filtered to your project type, with cross-references and Tier-1 source citations.

The 4-stage pipeline¶

┌──────────────────┐   ┌──────────────────┐   ┌──────────────────┐   ┌──────────────────┐
│  1. CONTEXT      │ → │  2. vibecoding-  │ → │  3. PLANNING     │ → │  4. EXECUTION    │
│  ENGINEERING     │   │     isold        │   │                  │   │                  │
│                  │   │                  │   │                  │   │                  │
│  WHAT to build?  │   │  WHAT MUST be    │   │  HOW to build?   │   │  Build it.       │
│  For whom?       │   │  in the plan?    │   │  (Phases, Tasks, │   │                  │
│  Which stack?    │   │  (Senior team    │   │   Architecture)  │   │                  │
│  Which scale?    │   │   requirements)  │   │                  │   │                  │
└──────────────────┘   └──────────────────┘   └──────────────────┘   └──────────────────┘
   PRD / Vision Doc   →   PLAN-REQUIREMENTS   →   Tech Plan          →   Code
   User Stories           .md (universal)         (any AI tool)
   Tech Stack             + Cursor Rules
   Constraints            + Claude Code Skill
                          + Cline Rules
                          + Aider Conventions
                          + Plain Prompt Template

We sit between context (what you're building) and plan (how it gets built) and ensure your plan covers what a senior team would have demanded. We don't replace your planning tool. We feed it.

A rendered SVG version of this diagram lives at assets/4-stage-pipeline.svg if your viewer supports it.

Where to go next¶

Quick Start :material-rocket-launch: Browse Profiles :material-folder-multiple-outline: View Examples :material-book-open-page-variant:

What's inside (V1.0)¶

Surface	Count	Where
Mandatory items	314	`core/roles/` + `core/PLAN-REQUIREMENTS.template.md`
Senior-engineering roles	7	`core/roles/{architect,backend,database,security,devops,sre,qa}.md`
Project profiles	6	`core/profiles/`
Tool integrations	5	`integrations/{cursor,claude-code-skill,cline,aider,prompt-templates}/`
Tier-1 primary citations	~970	Distributed across role files; >=2 per item
Cross-cutting anti-patterns	58	`core/anti-patterns.md`
Worked examples	6	`examples/`
Methodology	1	`methodology/brain-worker-orchestration.md`
Numerical Contracts (binding)	8	`core/_conflict-resolutions.md`

Three postmortems anchor the bet¶

Knight Capital lost $440M in 45 minutes because they shipped without a runtime kill switch. (Architect Item 39 — kill-switch / runtime feature-flag commitment.)
Cloudflare went down globally for 30 minutes in 2019 because one regex had no resource limit. (Backend Item 38 — ReDoS / catastrophic-backtracking protection.)
GitLab almost lost everything in 2017 because all 5 backup mechanisms had silently failed. (Database Item 21 + DevOps Item 26 — verified-restore drill.)

Each of those is a single item in the canonical source. There are 311 more.

We refuse to wait for the postmortem to learn what was missing.

What this is NOT¶

It IS	It IS NOT
A pre-planning knowledge injection layer	A coding agent
Stage 2 of a 4-stage pipeline	A planning tool (we feed planning tools, we don't replace them)
314 Mandatory items across 7 roles	A linter / static analyzer (we work before code is written)
6 project profiles, 184-264 items per project	A code-review tool
Tool-agnostic — Cursor, Claude Code, Cline, Aider, ChatGPT, plain Markdown	A "rules pack" with single-file opinions
MIT-licensed, free forever	A waitlist or pre-launch placeholder

Maintained by @0zoriginals under MIT. Solo maintainer, best-effort response, community PRs welcome.