BuildAgent World
Build · HIPM

HIPM

START
PLAN
DO
REVIEW
GOAL

HIPM is a team framework; my part is the design, the web UI, and the AI generators that produce its artefacts. A holographic phase wheel in R3F, a 2D-canvas particle field, the converging-orbs landing, glassmorphism throughout — all in Three.js + R3F + Framer Motion, hand-built rather than assembled from a kit.

Three.jsFramer MotionR3FGlassmorphism
Scroll
Four Interaction Experiments

Four ways to combine Three.js, R3F, and Framer Motion. Click any card to try it.

↓ Click to interact · move mouse away or scroll to release
Landing Animationlive

HIPM

Projektmanagement. Neu gedacht.

Landing Animation
Drei Welten. Ein Framework.live

Three Worlds. One Framework.

HIPM unites the strengths — and dissolves the weaknesses.

Drei Welten. Ein Framework.
5-Phasen-Modelllive

The 5-Phase Model

Click a phase for details

1
STARTWhy this project?

Create clarity, secure the mandate

2
PLANHow do we deliver this?

Turn orientation into commitment

3
DOWhat do we deliver — and when?

Deliver results, create value

4
REVIEWWhat have we learned?

Look honestly, get better

5
GOALDid we hit the goal?

Secure results, close the project

5-Phasen-Modell
Quality Gateslive

Quality through deliberate decision points.

Five gates safeguard progress — click a gate for details

Blue GateSTART

Can we start this project in a structured way?

Criteria

  • AProject purpose is clear, sponsor legitimized
  • BRoles named, escalation paths clear
  • CStakeholders identified, interests known
  • DApproach framework defined
  • EAI role documented

Abort

No sponsor or unresolvable goal conflicts

Fallback

Conflicting goals or unstable governance

Green GatePLAN

Do we have a resilient framework for execution?

Criteria

  • ADeliverables defined and prioritized
  • BExecution model set
  • CDecision paths documented
  • DTimeframe and dependencies known
  • EAI usage defined with clear boundaries

Abort

No viable execution perspective

Fallback

Open goal or governance questions

Orange GateDO

Have we demonstrably created value?

Criteria

  • AIncrements produced and accepted
  • BProgress documented, deviations justified
  • CRisk register updated
  • DStakeholders informed regularly
  • EAI outputs documented, human reviewed

Abort

Project goal no longer achievable

Fallback

Scope change requires replanning

Red GateREVIEW

Have we reflected honestly and drawn consequences?

Criteria

  • AResults checked against criteria
  • BRetrospective held, actions formulated
  • CLessons learned captured
  • DAI effectiveness analyzed
  • EPlan changes documented and released

Abort

Results fundamentally miss the objective

Fallback

Rework needed or planning errors identified

Purple GateGOAL

Is the project fully closed?

Criteria

  • ASponsor confirms goal attainment
  • BResults handed over, responsibility clarified
  • CLessons learned finalized and accessible
  • DAI tools handed over or shut down
  • EResources released, portfolio updated

Abort

Final gate — only fallback possible

Fallback

Handover failed or insights missing

Quality Gates
Next.js
Three.js
R3F
Framer Motion
Tailwind
Supabase
0

Standards Integrated

0+

Interactive Components

0

Quality Gates

0

Defined Roles

The phase wheel, up close

Holographic materials, physics-based interactions, scroll-driven animation. I wrote the shaders and the geometry myself; nothing here came out of a UI kit.

WheelSegment.tsx
// Each phase is an extruded arc with a custom holo shader —
// fresnel rim glow, scanlines, flicker, all driven by uTime.
const geometry = useMemo(
  () => createArcGeometry(
    WHEEL_INNER_RADIUS, WHEEL_OUTER_RADIUS,
    start, end, WHEEL_DEPTH,
  ),
  [start, end],
);

return (
  <mesh geometry={geometry}>
    <HoloMaterial
      color={PHASE_COLORS_HEX[phaseName]}
      opacity={isActive ? 0.55 : 0.3}
      glowIntensity={isHovered ? 1.6 : 1.0}
    />
  </mesh>
);

The generators

Four AI generators cover the START phase — Projektsteckbrief, Stakeholder-Analyse, Risikoübersicht, Kommunikationsplan — each wired to Claude via the Anthropic API. Claude drafts, leaves typed Rückfragenwhere the project is unclear, then finalises against your answers and scores a Reifegrad. The system prompt makes it a challenger, not a writer: it won't close a gap on its own.

prompts/projektsteckbrief.ts
// SCHRITT 1 — KI DENKT. Challenger-Mode, kein Schreiber.
export const SCHRITT_1_SYSTEM = `Du bist der Artificial Master
im HIPM-Framework (Hybrid Intelligent Project Management).
Deine Aufgabe: Aus unstrukturierten Projekt-Informationen einen
detaillierten Projektsteckbrief-ENTWURF generieren.

Du bist ein Challenger — kein Schreiber:
- Strukturiere, was der Mensch liefert
- Finde Lücken und Widersprüche
- Stelle unbequeme Rückfragen
- Schließe KEINE Lücken selbst — mache sie sichtbar

HIPM basiert auf ISO 21500, PMBOK, ICB 4, ISO 31000 und DIN 69901,
kombiniert mit Scrum-DNA und ist auf KMU zugeschnitten.`;

The generator in the browser

hipm.tools/generatoren/start/projektsteckbrief

Projektsteckbrief

Vollständiger Steckbrief nach HIPM-Standard

1

KI DENKT

Generiert Entwurf + Rückfragen

2

MENSCH LENKT

Beantworte Rückfragen, korrigiere

3

KI FINALISIERT

Überarbeitung + Konsistenzprüfung + Reifegrad

Was ist das Problem oder die Chance?Was soll konkret erreicht werden?Wer ist beteiligt (Rollen, Teams)?Budget und Zeitrahmen?Wo soll KI helfen?
Beschreibe dein Projekt: Was ist das Problem? Was soll erreicht werden? Wer ist beteiligt? Budget, Zeitrahmen?
Beispiel laden
Datei laden
Generieren

Gespeicherte Pläne (1)

HIPM_Steckbrief_V002

10.03.2026, 18:45

AnzeigenÜberarbeitenExportLöschen

Behind the UI

Supabase handles auth and plan storage with row-level security, so artifacts belong to an account, not a browser tab. Drafts save while they stream; plans export to Markdown, PDF, or DOCX. A closed tab never costs you anything.

Where I drew the lines

  • R3F instead of Spline

    Spline would have run in an hour, but it can't do custom shaders or scroll-coupled state. R3F gives me full Three.js inside the React lifecycle.

  • I scrapped the 3D hero

    I built an R3F convergence hero first — five orbs spiralling into the title with shockwave and bloom — then threw it out. Too subtle for the code it cost. The 2D Canvas field landed harder at a fraction of the weight.

  • Live materials, not pre-rendered images

    Transmission and metalness react to phase color and lighting in real time. Baking images would have meant rendering every phase against every theme up front, with nothing left to interact with.

  • The LLM challenges, it doesn't write

    Left alone, the model fills gaps with plausible but unchecked text. Telling it plainly that it isn't a writer forces the follow-up questions that make the output worth trusting.