Research Methodology · Aladdin V2

Old Lamp,
New Spirit

Aladdin V2 is the engineering method distilled from MRRC: IT veteran judgment awakened by vibe coding and agentic engineering, then disciplined by FDE, SDD, harness, and product leverage.

3Layer Loop
1Boundary Card
5Business Tools
Reuse Leverage

What is Aladdin V2?

A software engineering method for turning veteran experience into reusable agentic engineering assets.

Aladdin V2 = 老灯新灵 + FDE Cycle + SDD Spine + Agentic Harness + Product Leverage
老灯新灵 = Veteran Judgment + Vibe Coding + Field Truth
Veteran judgment defines direction and constraints; vibe coding releases execution speed; field truth keeps both honest.
In the Aladdin story, an old lamp looks ordinary until it is rubbed and the sealed spirit is released. In software engineering, the old lamp is the IT veteran's accumulated judgment, often sealed inside personal experience. The new spirit is the execution power released by vibe coding and AI agents. Aladdin V2 turns that release into a disciplined engineering system.

Three Layers, One Loop

The seven-layer model collapses to one operational loop. Each layer answers a single question, carries one constraint, and buys one effect.

L1 · Constrain

Set the Boundary

Who: veteran + field truth

Decide what to do, what not to do, what must hold — before any code.

ConstraintWrite the Boundary Card first; over 10 lines means it isn't clear yet.
EffectTurns AI's fast guessing into fast execution.
L2 · Execute

Run Inside It

Who: AI / agent, inside the boundary

Vibe coding for speed; the FDE rhythm (Echo→Delta→Product) for direction.

ConstraintValidate the biggest uncertainty first; converge in 3–5 rounds; demo each round.
EffectThe speed the boundary just bought you.
L3 · Capture

Turn It Into Asset

Who: human + AI together

SDD, harness, ADR, postmortem — the carriers that outlive this cycle.

ConstraintEvery cycle leaves at least one reusable asset, or it isn't done.
EffectNext time is cheaper — that is leverage.
Boundary → Speed → Asset → (cheaper) Boundary

One Turn of the Loop

Constrain, execute, capture — then the next boundary is cheaper to draw.

L1Set the Boundary
L2Execute Inside It
L3Capture the Asset

Why MRRC is the Proof Ground

Real Hardware

Radios, USB audio devices, ATR-1000 tuner, serial control, RTL-SDR, TLS, and local system dependencies prevent purely imaginary solutions.

Real-Time Constraints

PTT timing, TX/RX audio queues, AudioWorklet buffering, Opus frame alignment, and WebSocket scheduling expose real engineering risk.

Complete Architecture Memory

FDE retrospectives, SDD 14 chapters, DESIGN docs, postmortems, AGENTS, and agentic plans show how knowledge becomes reusable.

Repo-Local Skill

The methodology is operationalized as an OpenCode skill: .opencode/skills/aladdin-v2/SKILL.md.

The Aladdin V2 Rules

1

L1Write the boundary before the code. No boundary, no agent.

2

L1Field truth outranks opinion; speed lives only inside the boundary.

3

L2Delta code is disposable; the captured asset is what you keep.

4

L3Every cycle must make the next one cheaper — or it isn't finished.

5

A method must mark its own range. Outside it, bring your own equipment.

An Honest Look: Uneven Maturity

A method is only trustworthy once it marks its own boundaries. Here is where Aladdin V2 is strong, and where it has never been pressure-tested.

Aladdin V2 is not an immature methodology — it is a methodology of highly uneven maturity. On the dimensions a solo hardware project hammered repeatedly (execution constraints, knowledge capture, architecture memory) it reaches rare engineering density. On the dimensions that project never stressed (multi-person judgment, authority correction, budget & safety, high-frequency feedback) it is near-blank. The blanks are not design failures — they are an unexplored range.

Verified by real artifacts

  • Executable plans with exact line numbers, code blocks, py_compile checks, expected output, even pre-written commits.
  • SDD spine: all 14 chapters present, not an empty shell.
  • Memory harness: multiple postmortems hardened into AGENTS.md guardrails.
  • A triggerable OpenCode skill with real phase detection, not motivational prose.

Never pressure-tested

  • Team dimension: no arbitration when two veterans' judgments conflict; no cold-start path when there is no veteran.
  • Correction loop: veteran judgment is the top authority with no falsification entry when it is itself wrong.
  • Cost & safety: token blow-ups, agent touching production, supply-chain risk — unmentioned.
  • Feedback latency: never modeled as a first-class variable.
The correct stance is not "this method is flawed, so discount it." It is: within its range it is precision-built; outside that range you must bring your own equipment — and the only real omission of the original author was failing to mark the range boundary. The toolkit below is that equipment.

Five Tools to Take Into New Waters

Pick what you need. Lowest barrier: just use the Boundary Card. The rest scale with team maturity.

1 · Boundary Card

Under 10 lines, written before anyone (or any agent) starts: goal, what's out of scope, what must hold, the 3 things most likely to be guessed wrong, how failure is detected. Highest ROI, lowest migration cost.

2 · Seven Harnesses

The original six (Context / Spec / Plan / Verification / Runtime / Memory) plus a new Safety/Cost harness: what an agent may touch, the budget ceiling, the human-confirm points before production.

3 · Three Hard Metrics

Replace qualitative scores with falsifiable trend lines: handoff exploration time, repeat-incident count, marginal cost of a similar feature. Leverage is real only if these curves bend down.

4 · Correction Loop

At each key decision, one lightweight challenge: a peer's reverse argument, an "if the veteran is wrong" stress test, or an evidence-source audit. A falsification entry for the single judgment source.

5 · Range Annotation

Mark applicability before adopting: stable domain, concentrated judgment, long feedback cycle = high fit. High-churn discovery or regulation-over-field domains need the FDE rhythm rewritten.

Each tool patches one blind spot

Correction loop ↔ single-hero gap · Range annotation ↔ n=1 overfit · Hard metrics ↔ qualitative self-comfort · Safety/Cost harness ↔ the cost & safety silence.

Read the Methodology

Aladdin V2 is maintained in the MRRC repository under docs/aldv2/, together with the OpenCode skill definition.

View Docs