Learning & research

Program Synthesis

Explore program generation and evaluation in a meta-agentic research workflow.

OPEN THE COMPLETE COMMAND DECKS

Explore the model in your browser.

Read-only dashboards with energy, governance, stress scenarios and preserved diagrams. No wallet or installation required; all values come from recorded simulations.

A CLOSER LOOK

How does a mission become an evaluated program candidate?

For operators and developers: search bounded programs, challenge candidates with a separate checker, prepare USDC work orders and inspect both original synthesis engines.

REAL REPOSITORY MATERIAL

Inspect. Understand. Reproduce.

Reading the exact source at revision 5b4cebb3.
This browser inspection does not execute the demo.

demo/Meta-Agentic-Program-Synthesis-v0/config/mission.meta-agentic-program-synthesis.json

Select a walkthrough step to explore its source.

Full source text
{
  "meta": {
    "version": "0.1.0",
    "title": "Meta-Agentic Program Synthesis Sovereign Mission",
    "subtitle": "Evolutionary Intelligence Forge",
    "description": "Deterministic rehearsal proving that AGI Jobs v0 (v2) lets a non-technical owner conjure, govern, and redeploy an autonomous meta-agent that designs and verifies production-ready code modules in minutes.",
    "ownerAddress": "0x1111111111111111111111111111111111111111",
    "treasuryAddress": "0x2222222222222222222222222222222222222222",
    "timelockSeconds": 604800,
    "governance": {
      "council": [
        "sovereign.validator.eth",
        "thermostat.guardian.eth",
        "alpha.operator.eth"
      ],
      "sentinels": [
        "sentinel.quantum",
        "sentinel.reward-engine",
        "sentinel.timelock",
        "sentinel.contract-size"
      ],
      "ownerScripts": [
        "npm run owner:command-center",
        "npm run owner:atlas",
        "npm run owner:system-pause",
        "npm run owner:upgrade-status",
        "npm run owner:change-ticket"
      ]
    }
  },
  "parameters": {
    "seed": 133742,
    "generations": 9,
    "populationSize": 36,
    "eliteCount": 6,
    "crossoverRate": 0.45,
    "mutationRate": 0.32,
    "maxOperations": 6,
    "energyBudget": 480,
    "successThreshold": 0.965,
    "noveltyTarget": 0.72
  },
  "qualityDiversity": {
    "complexityBuckets": [1, 2, 3, 4, 5, 6],
    "noveltyBuckets": [0.1, 0.25, 0.5, 0.75, 0.9, 1.0],
    "energyBuckets": [60, 120, 180, 240, 360, 480]
  },
  "tasks": [
    {
      "id": "arc-sentinel",
      "label": "ARC Sentinel Edge Lift",
      "narrative": "Detect latent pixel edges, amplify discovery, and route high-confidence transformations back into the shared skill graph.",
      "mode": "vector",
      "pipelineHint": ["difference", "threshold", "scale"],
      "constraints": {
        "maxOperations": 5,
        "preferredOperations": ["difference", "threshold", "scale", "mirror"],
        "expectedEnergy": 96
      },
      "examples": [
        {
          "label": "Dual-edge pattern",
          "input": [0, 0, 1, 1, 0, 0],
          "expected": [0, 0, 3, 0, 3, 0]
        },
        {
          "label": "Shoulder uplift",
          "input": [0, 1, 1, 1, 0, 0],
          "expected": [0, 3, 0, 0, 3, 0]
        },
        {
          "label": "Symmetric shell",
          "input": [1, 1, 0, 0, 1, 1],
          "expected": [0, 0, 3, 0, 3, 0]
        }
      ],
      "owner": {
        "jobId": "ARC-SYN-001",
        "stake": 120000,
        "reward": 420000,
        "thermodynamicTarget": 88.0
      }
    },
    {
      "id": "ledger-harmonics",
      "label": "Ledger Harmonics Sequencer",
      "narrative": "Absorb asynchronous cash-flow deltas, stabilise them against the thermodynamic ledger, and emit rebalanced incentive curves for validator coalitions.",
      "mode": "vector",
      "pipelineHint": ["cumulative", "mod", "offset"],
      "constraints": {
        "maxOperations": 5,
        "preferredOperations": ["cumulative", "mod", "offset", "scale"],
        "expectedEnergy": 128
      },
      "examples": [
        {
          "label": "Staggered settlement",
          "input": [5, -3, 4, -2],
          "expected": [2, 4, 3, 6]
        },
        {
          "label": "Equilibrium climb",
          "input": [1, 1, 1, 1],
          "expected": [3, 4, 5, 6]
        },
        {
          "label": "Validator correction",
          "input": [7, -4, 2, -3],
          "expected": [4, 5, 2, 4]
        }
      ],
      "owner": {
        "jobId": "LEDGER-SYN-007",
        "stake": 155000,
        "reward": 525000,
        "thermodynamicTarget": 64.0
      }
    },
    {
      "id": "nova-weave",
      "label": "Nova Weave Sequence Optimiser",
      "narrative": "Elevate alpha sequences into deterministic blueprints that compile into on-chain production payloads with zero manual edits.",
      "mode": "vector",
      "pipelineHint": ["power", "offset", "scale"],
      "constraints": {
        "maxOperations": 6,
        "preferredOperations": ["power", "offset", "scale", "mod"],
        "expectedEnergy": 144
      },
      "examples": [
        {
          "label": "Prime harmonic",
          "input": [2, 3, 4],
          "expected": [10, 20, 34]
        },
        {
          "label": "Token feedback",
          "input": [1, 5, 2],
          "expected": [4, 52, 10]
        },
        {
          "label": "Validator lattice",
          "input": [3, 6, 1],
          "expected": [20, 74, 4]
        }
      ],
      "owner": {
        "jobId": "NOVA-SYN-009",
        "stake": 165000,
        "reward": 610000,
        "thermodynamicTarget": 68.0
      }
    }
  ],
  "ci": {
    "workflow": "ci (v2)",
    "requiredJobs": [
      { "id": "lint", "name": "Lint & static checks" },
      { "id": "tests", "name": "Tests" },
      { "id": "foundry", "name": "Foundry" },
      { "id": "coverage", "name": "Coverage thresholds" }
    ],
    "minCoverage": 90,
    "concurrency": "ci-${{ github.workflow }}-${{ github.ref }}"
  },
  "ownerControls": {
    "capabilities": [
      {
        "category": "Emergency Pause",
        "label": "Circuit breaker engage",
        "command": "npm run owner:system-pause -- --action pause",
        "verification": "npm run owner:system-pause -- --action status"
      },
      {
        "category": "Thermostat",
        "label": "Recalibrate reward engine temperature",
        "command": "npm run thermostat:update -- --mission demo/Meta-Agentic-Program-Synthesis-v0/config/mission.meta-agentic-program-synthesis.json",
        "verification": "npm run thermodynamics:report"
      },
      {
        "category": "Upgrade",
        "label": "Queue sovereign upgrade",
        "command": "npm run owner:upgrade -- --mission demo/Meta-Agentic-Program-Synthesis-v0/config/mission.meta-agentic-program-synthesis.json",
        "verification": "npm run owner:upgrade-status"
      },
      {
        "category": "Treasury",
        "label": "Mirror treasury share",
        "command": "npm run reward-engine:update -- --mission demo/Meta-Agentic-Program-Synthesis-v0/config/mission.meta-agentic-program-synthesis.json",
        "verification": "npm run reward-engine:report"
      },
      {
        "category": "Compliance",
        "label": "Refresh compliance dossier",
        "command": "npm run owner:compliance-report",
        "verification": "npm run owner:doctor"
      }
    ]
  }
}

SHA-256 ecc542c8f48a08de609cb7d24d6bc1f502af6ce2b763c707b0265d818cb2d515

FROM READING TO A REPRODUCIBLE RUN

Try the selected path.

Node.js workspace

Run from the repository root after nvm install, nvm use and npm ci. Use the versions pinned in .nvmrc and package.json. Commands below describe the selected path; optional app/server packages can have their own locked dependencies.

Dependency files for this demo and its variants (1)
Complete environment setup ↗
SELECTED EXECUTION PATH
npm run demo:program-synthesis:site

What you should observe

A complete offline website in build/program-synthesis, including the executable lab, work-order exports, regenerated Python command theatre and TypeScript dashboard.

The source inspector above reads bundled repository material. Local commands run separately on your computer. Recorded examples may contain historical timestamps, placeholders and simulated metrics.

MAKE IT YOUR OWN

One useful experiment.

Search a program, export a wrong candidate, and run the Python checker. Set reviewer capacity to zero and inspect why the draft work order is held.

THE SYSTEM, MADE VISIBLE

Architecture & relationships

Architecture diagram · source preserved below
View original Mermaid source
flowchart TD
    Objective["Authorized objective"] --> Admission["Scope, budget and reviewer capacity"]
    Admission --> Search["Bounded synthesis or commissioned worker"]
    Search --> Evidence["Candidate and source-bound artifacts"]
    Evidence --> Check["Task-specific acceptance checks"]
    Check -->|Fails| Correction["Correct or reject"]
    Correction --> Search
    Check -->|Passes| Review["Independent review and buyer acceptance"]
    Review --> Settlement["Separately authorized settlement"]
    Settlement --> Learning["Accepted-outcome learning"]
    Learning --> Admission

WHEN SOMETHING DOESN’T MATCH

Troubleshooting

A command or service fails

Start at the first error, confirm the working directory and pinned toolchain, then follow the linked component guide. Optional packages and provider services have separate prerequisites.

A report looks successful but lacks evidence

Check its source revision, configuration, timestamps and underlying events. Historical fixtures and generated plans do not prove a fresh execution.

TRACE THE CHECKS

Verification & next steps

13 tracked test source files are available in this directory. Inspect the tests and their environment before choosing a suite; file counts do not establish test results.

Browse the test sources

For live commissioning, consult the production readiness record.

EVERY VARIANT, PRESERVED

Complete document library

REPRODUCE & INSPECT

Registered commands

Run commands from the repository root after following this demo's guide. Network and owner actions require their documented setup.

npm run demo:meta-agentic-program-synthesis
npm run demo:meta-agentic-program-synthesis:briefing
npm run demo:meta-agentic-program-synthesis:full
npm run demo:program-synthesis:site
npm run demo:program-synthesis:lab
npm run demo:program-synthesis:test
npm run demo:program-synthesis:qa
npm run demo:program-synthesis:worker

Full command catalog and troubleshooting ↗

ORIGINAL DIRECTORYdemo/Meta-Agentic-Program-Synthesis-v0View on GitHub ↗