ContextQB Methodology

Architectural thinking for AI-native builders.

What's here

Principles, playbooks, audits, and prompts that teach systems thinking for AI-assisted software development; the context.qb format spec; the MCP Worker that backs mcp.contextqb.com; and the website's markdown copy.

content/
  principles/    ← Core architecture principles                    (CC BY-SA 4.0)
  playbooks/     ← Step-by-step workflows                          (CC BY-SA 4.0)
  audits/        ← Agent audit templates                           (CC BY-SA 4.0)
  prompts/       ← Reusable agent prompts                          (CC BY-SA 4.0)
  guides/        ← Operator guides                                 (CC BY-SA 4.0)
format/
  SPEC.md        ← The context.qb file format specification        (CC BY 4.0)
  examples/      ← Example context.qb files                        (CC BY 4.0)
apps/
  mcp/           ← MCP Worker source (Cloudflare Workers + D1)     (MIT)
  web/
    content/     ← Website page copy + briefings                   (CC BY-SA 4.0)
docs/
  mcp-setup.md   ← How to add the MCP to Cursor / Claude / etc.

Anything not listed here (application UI source, course platform, strategy, internal docs) lives in the private upstream and is not published.

How to use it

Read on the web

Browse the full methodology at contextqb.com.

Connect your AI agent

Give your AI agent direct access to all methodology content via MCP. See docs/mcp-setup.md.

Use the content directly

All methodology content is licensed CC BY-SA 4.0. You can read, copy, share, and adapt it with attribution. Derivatives must use the same license.

The context.qb format specification is CC BY 4.0.

Courses

Structured learning at courses.contextqb.com.

License

See LICENSE for details.

---

AI can write code. Someone still has to think.