mORMot Open Source

flydev

Member

From: France

Registered: 2020-11-27

Posts: 172

Website

ai superpowers for mormot2

I forked obra/superpowers into a Pascal/Delphi+FPC variant specialized for mORMot2 - so it work for claude, codex or gemini, etc.

Repo: https://github.com/flydev-fr/mormot2-superpowers

What it adds:
  - 10 domain skills that route prompts to real mORMot 2 idioms, and a Pascal-aware code-reviewer agent.
  - /mormot2-doc <topic> that resolves topic names ("torm", "rest", "ecc", "ddd", "logging", ...) or chapter numbers to local SAD chapter excerpts (to be improved / automated / reviewed) and will pull @zen010101 wiki soon.
  - /delphi-build and /fpc-build that wrap dcc32/dcc64/MSBuild and fpc/lazbuild with mORMot 2 search paths injected automatically.
  - TSynTestCase-only test discipline. /mormot2-test does not run DUnitX/FPCUnit/TestInsight (deliberate).
  - Cross-platform: Windows (Delphi + FPC), Linux, macOS.

Quick example:

> Add a TOrm class for a User with email and role fields
[Skill: mormot2-orm loaded]
[generates a TOrmUser with RawUtf8 fields, stored AS_UNIQUE on email,
the right index attribute, and tells you how to register it on a TOrmModel]

You can tests from your shell:

pwsh -File tests/run-quick.ps1

Note: the upstream methodology layer (planning, TDD, subagent-driven development) is preserved unchanged; the Pascal/mORMot layer is the new bit.

Feedback welcome - especially on missing idioms, wrong class names, or trigger-eval misroutes. Open an issue or reply here

@zen010101

Last edited by flydev (2026-04-26 12:08:21)

ab

Administrator

From: France

Registered: 2010-06-21

Posts: 15,475

Website

Re: ai superpowers for mormot2

Sounds amazing, but honestly you lost me here a little.
Even the readme is a bit short to understand it fully.

Can you write/generate some more documentation/content about the purpose of this project, and show some input/output on the purpose of this tool?
I could eventually add it to the Synopse blog, for sure...

flydev

Member

From: France

Registered: 2020-11-27

Posts: 172

Website

Re: ai superpowers for mormot2

Ok, let me write a better README, and I'll take the time to make a screencast or video comparing both runs: without, and with the plugin.

I assume you’re not familiar with the superpowers plugin. Do you use claude-code yourself?

zen010101

Member

Registered: 2024-06-15

Posts: 173

Re: ai superpowers for mormot2

Wow, these 10 practical skills are truly excellent and highly useful. They can effectively make AI-generated programs more consistent and compliant with the mORMot-Style specifications! I sincerely appreciate the hard work and outstanding development efforts from flydev. When it comes to the overall process framework design and construction, I personally highly recommend the BMAD Method. That’s right, I have fully migrated from superpower to this framework long ago. Although its overall structure is more complex, it is completely worthwhile once you become proficient in using it.

flydev

Member

From: France

Registered: 2020-11-27

Posts: 172

Website

Re: ai superpowers for mormot2

Thanks @zen. I remember now that you mentioned the BMAD method in another thread. I also remember that, after reading the README, I wanted to actively test it but then I completely forgot to do it!

Thanks for the reminder - I'll definitely give it a real try, and maybe eventually publish something specialized in the same way I did for superpowers.

flydev

Member

From: France

Registered: 2020-11-27

Posts: 172

Website

Re: ai superpowers for mormot2

hey @zen - small question, once the workflow architecture is complete, did you scaffold the project with vanilla claude agents by just running /bmad-create-story ?

To test BMAD this afternoon, I have reset my claude installation, installed BMAD and mormot2-superpowers plugin. Then ran BMAD on a new project and once BMAD finished, I ran /bmad-help and then it started using mormot2-superpowers on the BMAD architecture document.

Architecture workflow complete. All 8 steps signed off, frontmatter sealed: stepsCompleted: [1..8],
  status: complete, completedAt: 2026-05-14.

  What we built together

  A 30 + KB architecture document at _bmad-output/planning-artifacts/architecture.md that takes the PRD's 55 FRs and ~45 NFRs and lays them out as a concrete, implementable system:
  ...
 
Persistent memory written this session

  - feedback-fpc-toolchain.md — FPC 3.2.3-fixed path
  - feedback-no-ifdef-fpc.md — mORMot 2 abstracts compiler divergence
  - feedback-mormot2-tests-as-docs.md — test/ + ex/ as the canonical reference
  - feedback-mormot2-track-master.md — track master, not stable tags; local clone is truth
  - project-license-deprio.md — license work deprioritized for v1
  - reference-mormot-soa-blogs.md — Synopse SOA blog + forum #3621 URLs

  Next concrete steps

  1. Run Skill: mormot2-superpowers:mormot2-init to scaffold .claude/mormot2.config.json with the FPC 3.2.3-fixed path
  and the mORMot 2 + mormot-extensions search paths.
  ...

btw - started at 12:45 and it just finished (almost 10 hours non-stop) - all tests are greens and I didn't write a single line of code written (sadly..) and still not reviewed the code but I must admit BMAD is powerful (I was using cavekit blueprints which is a bit similar but not powerful as BMAD is from what I feel).