ADD(A, 2)Increase a value
Add 2 to A. With A=8, the new value is A=10. It fails if the bounded value overflows.
// INTERACTIVE EXECUTION RESEARCH
BUILD THE BLOCK.
TEST THE RULES_
Three hands-on puzzles with missing operations, a deliberate distractor and a decision about ordering. Learn the rules first, then solve them in the browser.
BROWSER-ONLY MODELNO WALLETNO REAL ASSETS
READ THE OPERATOR MANUAL FIRST
01 / START HERE: OPERATOR MANUALREAD BEFORE EDITING
// THE MISSION BRIEFING
Understand the rules.
Solve the puzzle.
This playground lets you arrange simple instructions that change pretend balances. It checks your result with a browser-only model of PoUT's isolated CWC research. Each challenge has a specific goal — not every valid calculation is a winning solution.
First-time visitor? Follow the four steps on the right. If an operation is unfamiliar, use the examples below before pressing RUN.
01
Choose a challenge
Parallel Protocol asks you to finish a three-operation plan. Atomic Shield has a distractor to remove and a missing transfer to add. Order Matters starts with the wrong candidate first. Free Build has no goal.
02
Read the objective and initial values
Look at Your Objective, Starting Puzzle and the registers A, B, R, X and Y. Those numbers are pretend balances. Missions have a prescribed starting state; changing it may make the puzzle unwinnable. RESET CHALLENGE restores the starting puzzle, not your earned badges.
03
Construct your candidate
Use the Operation Queue to inspect existing steps. The × deletes a step, arrows change its position, and the Queue New Operation form adds instructions. Select operation, key, optional destination and amount. The graph shows which steps can share a wave.
04
Run, read both results, adjust
Press RUN EXPERIMENT. Execution Status describes whether the candidate is valid or rejected. Mission Status tells you if you solved the actual goal. Valid can still mean not solved. For Atomic Shield, a carefully targeted rejection is the correct win.
INSTRUCTION SET / BEGINNER REFERENCE
The four operations explained
Independent practice example: start with A=8 and B=5. These examples do not reveal any mission solution.
MOVE(A → B, 3)Transfer a value
Take 3 from A and give 3 to B. With A=8, B=5, the new values are A=5 and B=8. It fails if A has less than 3.
SET(A, 4)Replace a value
Set A directly to 4, regardless of its previous value. This changes a synthetic counter; it does not mint coins.
ASSERT_EQUALS(B, 5)Check without changing
Accept this check only if B is exactly 5 at that point in the sequence. The register does not change.
What is a dependency wave? +
The model groups operations that do not conflict into the same wave. For example, ADD(A,2) and ADD(B,3) access different registers and may be evaluated together. Two instructions touching A may require separate waves. The order you place operations in the queue matters. Worker lanes 1/2/4 only change the visualization; this site does not run Rust worker threads or measure speed.
Why is an invalid candidate sometimes a mission win? +
Atomic means all-or-nothing. When a candidate tries to spend more than it has, the correct safe outcome is to reject the entire candidate, not to apply half of its changes. The mission checks that you caused the intended overdraw, not just any unrelated error. In the separate-candidate experiment, the first valid candidate may commit and the second can be rejected against its changed state.
What do the green and amber messages mean? +
MISSION COMPLETED (green) means the exact puzzle goal was met. MISSION NOT COMPLETED (amber) means you have more work to do, even if the operations were valid. CANDIDATE ACCEPTED means only that the model permits this sequence. INVALID CANDIDATE REJECTED can be the correct safety behavior. An empty queue may be technically accepted but is never a mission win.
How does saving work? Can I lose progress? +
Only completed mission badges are stored in localStorage on this browser and website address. No account, cloud sync, wallet, server or saved transaction history. Clearing website data (often shown alongside cookies) deletes progress. Clearing only the image cache normally does not. Private browsing and different browsers, devices or domains have separate progress. These are redesigned v0.4 challenges; badges from v0.3 are not counted toward the new puzzles. Use CLEAR BADGES if you want to start again.
Can I use this to test the actual PoUT blockchain? +
No. The Playground uses a JavaScript teaching model of selected Rust CWC finite-state rules, not the Rust executable and not a live PoUT node. Values are capped for the browser interface, with just five synthetic registers and twelve operations. Its results are not security audits or proof that a native consensus protocol is complete.
02 / EXPERIMENT TERMINALALL INPUTS ARE SYNTHETIC
Research Playground
Each mission starts unfinished. Follow the objective, complete the missing steps, then test your solution. Read the manual ↑
CHALLENGES SOLVED00 / 03LOCAL SAVE INITIALIZING
02B / EXECUTION VISUALIZER
Dependency graph
MODEL READY
Loading graph ...
Independent in same wave Dependency / next wave Candidate rejection ≠ network failure
MODEL SEMANTICS
Operations in one wave have no conflicting read/write accesses. All changes commit only if the complete candidate is valid.
[ INPUT → PLAN → VALIDATE ]
03A / TRANSACTION BUILDER
Operation queue
The starting queue is not the solution. Follow the mission objective, then add, remove or reorder steps.
INPUTS LOCK DURING RUN • NO NETWORK REQUESTS • NO REAL FUNDS
03B / EXECUTION OUTPUT
Mission & execution results
Press RUN EXPERIMENT to inspect the candidate.
pout@lab:~/cwc
> PoUT CWC browser model initialized.
> Choose a mission or compose your own workload.
> Awaiting local input_
LAST RUN: NOT EXECUTEDTRANSPORT: DISABLED
// TRUST BOUNDARIES
What this lab is not.
Not a live PoUT node
Runs a lightweight JavaScript model of documented CWC rules. It does not call the Rust executor, join a network, produce blocks or measure real throughput.
Not a security audit
Presets reflect documented finite test scenarios. Browser output cannot establish production correctness, scarcity of useful work or secure consensus.
Nothing is submitted
No wallet, token purchase, login, telemetry, backend, remote assets or third-party requests. Only three current-version completed-mission flags are saved locally on this origin; no workload history or personal data is transmitted.