Agents & applications

Meta-Agentic ALPHA AGI Jobs

Evaluate six delivery phases, inspect independent evidence and explore every preserved mission console.

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 meta-agentic loop turn useful project briefs into accountable work orders?

For operators and builders: qualify specialized workers, define concrete deliverables, reserve USDC budgets and reviewer time, and challenge evidence across six executable phases.

Evaluate the six-phase loop

Inventory twelve proposed projects, qualify from synthetic observations, route specialists and reserve budget and reviewer capacity. Five work orders fit the reference scenario; customer projects are not executed.

demo/Meta-Agentic-ALPHA-AGI-Jobs-v0/workbench/execute.mjs ↗

Operate and explore every version

Use the current admitted OpenClaw path or operator-led Work candidate flow, and preserve all twelve Python demonstrations and the classic TypeScript view.

demo/Meta-Agentic-ALPHA-AGI-Jobs-v0/RUNBOOK.md ↗

REAL REPOSITORY MATERIAL

Inspect. Understand. Reproduce.

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

Repository source5,023 bytesDownload source ↓View on GitHub ↗

demo/Meta-Agentic-ALPHA-AGI-Jobs-v0/workbench/execute.mjs

Select a walkthrough step to explore its source.

Full source text
import {
  validateScenario,
  amount,
  decimal,
  boundary,
  stages,
  canonical,
  sha256,
  json,
  makeTask,
  taskDigest,
} from './model.mjs';

export function evaluate(source) {
  validateScenario(source);
  const workers = source.workers.map((w) => ({
    ...w,
    qualified:
      w.reviewed >= source.minimumReviewed &&
      w.useful * 10000 >= w.reviewed * source.minimumUsefulBps,
  }));
  const routes = source.work.map((w) => ({
    workId: w.id,
    workerId:
      workers
        .filter((x) => x.skill === w.skill && x.qualified)
        .sort((a, b) => b.useful * a.reviewed - a.useful * b.reviewed)[0]?.id ??
      null,
    reviewPool: w.skill,
  }));
  const contracts = source.work.map((w, i) => ({
    workId: w.id,
    workerId: routes[i].workerId,
    deliverable: w.deliverable,
    acceptanceCriteria: w.acceptanceCriteria,
    rewardUsdc: decimal(amount(w.rewardUsdc)),
    reviewPool: w.skill,
    reviewMinutes: w.reviewMinutes,
  }));
  let remaining = amount(source.budgetUsdc) - amount(source.reserveUsdc),
    committed = 0n;
  const pools = source.reviewPools.map((p) => ({
    id: p.id,
    availableMinutes: p.minutes,
    usedMinutes: 0,
  }));
  const decisions = contracts.map((c) => {
    const pool = pools.find((p) => p.id === c.reviewPool),
      cost = amount(c.rewardUsdc);
    const reason = !c.workerId
      ? 'no-qualified-worker'
      : cost > remaining
      ? 'budget'
      : pool.availableMinutes - pool.usedMinutes < c.reviewMinutes
      ? 'review-capacity'
      : 'reserved';
    if (reason === 'reserved') {
      remaining -= cost;
      committed += cost;
      pool.usedMinutes += c.reviewMinutes;
    }
    return {
      workId: c.workId,
      status: reason === 'reserved' ? 'admitted' : 'deferred',
      reason,
    };
  });
  return {
    identify: {
      workIds: source.work.map((w) => w.id),
      count: source.work.length,
      plannedRewardsUsdc: decimal(
        source.work.reduce((n, w) => n + amount(w.rewardUsdc), 0n)
      ),
      reviewMinutes: source.work.reduce((n, w) => n + w.reviewMinutes, 0),
      ...boundary,
    },
    learn: {
      workers: workers.map((w) => ({
        id: w.id,
        reviewed: w.reviewed,
        useful: w.useful,
        usefulBps: w.reviewed ? Math.floor((w.useful * 10000) / w.reviewed) : 0,
        qualified: w.qualified,
      })),
      ...boundary,
    },
    think: { routes, ...boundary },
    design: { contracts, ...boundary },
    strategise: {
      decisions,
      committedUsdc: decimal(committed),
      unallocatedUsdc: decimal(amount(source.budgetUsdc) - committed),
      reviewPools: pools,
      ...boundary,
    },
    execute: {
      readyWorkIds: decisions
        .filter((x) => x.status === 'admitted')
        .map((x) => x.workId),
      deferredWorkIds: decisions
        .filter((x) => x.status === 'deferred')
        .map((x) => x.workId),
      committedUsdc: decimal(committed),
      unallocatedUsdc: decimal(amount(source.budgetUsdc) - committed),
      customerJobsCompleted: 0,
      actualPaidUsdc: '0.000000',
      evidenceClass: 'deterministic-simulation',
      ...boundary,
    },
  };
}
export async function execute(source) {
  const outputs = evaluate(source),
    results = [];
  for (const stage of stages) {
    for (const dep of stage.dependencies)
      if (!results.some((x) => x.stageId === dep))
        throw new Error('Missing phase dependency.');
    const content = json(outputs[stage.id]);
    results.push({
      stageId: stage.id,
      dependencies: stage.dependencies,
      taskSha256: await taskDigest(await makeTask(source, stage.id)),
      artifact: {
        name: stage.file,
        mediaType: 'application/json',
        content,
        bytes: new TextEncoder().encode(content).length,
        sha256: await sha256(content),
      },
    });
  }
  return {
    schemaVersion: 1,
    scenarioId: source.id,
    sourceSha256: await sha256(canonical(source)),
    evidenceClass: 'deterministic-simulation',
    providerCalls: 0,
    chainTransactions: 0,
    ...boundary,
    results,
  };
}
export function renderReport(bundle) {
  const d = JSON.parse(bundle.results.at(-1).artifact.content);
  return `# Meta-Agentic ALPHA — delivery evaluation\n\nSource SHA-256: ${
    bundle.sourceSha256
  }\n\n${d.readyWorkIds.length} review-ready work orders; ${
    d.deferredWorkIds.length
  } deferred. Planned commitment: ${
    d.committedUsdc
  } USDC. Unallocated (including reserve): ${
    d.unallocatedUsdc
  } USDC.\n\nReady: ${
    d.readyWorkIds.join(', ') || 'none'
  }.\n\nSix local evaluation phases completed. Proposed customer jobs completed: 0. Provider calls: 0. Transactions: 0. No production or settlement approval.\n\n${bundle.results
    .map(
      (r) => `- ${r.stageId}: ${r.artifact.name} — SHA-256 ${r.artifact.sha256}`
    )
    .join(
      '\n'
    )}\n\nA passing content check does not establish provider provenance, independent human acceptance, job fulfillment or payment authority.\n`;
}

SHA-256 fccebc1fcb0b70fa8a24dd7c6d1ec8f8378be370f9ff501631f67b7352365bf5

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.

Complete environment setup ↗
SELECTED EXECUTION PATH
npm run demo:meta-agentic-alpha:work

What you should observe

Six checked JSON artifacts, six exact task contracts and an evidence dossier. Reference result: five review-ready work orders, seven deferred, 23,000 USDC reserved and 7,000 unallocated. No provider calls, customer-job completion or payment.

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.

Evaluate the portfolio, test a rehashed wrong answer and set reviewer hours to zero. Explore the preserved consoles and distinguish their fixture scores from actual release CI.

THE SYSTEM, MADE VISIBLE

Architecture & relationships

Architecture diagram · source preserved below
View original Mermaid source
graph TD
    owner[Owner Multi-Sig\n0xF39Fd6e51aad88F6F4ce6aB8827279cffFb92266] --> orchestrator[Meta-Agentic Planner]
    orchestrator --> treasury[Treasury Manager]
    orchestrator --> governance[On-Chain Governance]
    treasury --> a2aBus
    governance --> a2aBus
    jobRegistry[AGI Job Registry\n2.2.0]
    treasuryManager[Autonomous Treasury Manager\n2.2.0]
    a2aBus[Agent-to-Agent Coordination Bus\n2.1.1]
    worldModel[MuZero-Grade World Model\n1.9.0]
    governance[Timelock + Multisig Controller\n2.0.3]
    execute-autonomy -->|leads| opp-supply-inversion
    think-orchestrator -->|supports| opp-supply-inversion
    identify-scout -->|supports| opp-supply-inversion
    strategy-supreme -->|supports| opp-supply-inversion
    learn-forge -->|supports| opp-supply-inversion
    design-savant -->|supports| opp-supply-inversion
    execute-autonomy -->|leads| opp-latent-alpha
    think-orchestrator -->|supports| opp-latent-alpha
    identify-scout -->|supports| opp-latent-alpha
    strategy-supreme -->|supports| opp-latent-alpha
    learn-forge -->|supports| opp-latent-alpha
    execute-autonomy -->|leads| opp-bio-discovery
    think-orchestrator -->|supports| opp-bio-discovery
    identify-scout -->|supports| opp-bio-discovery
    learn-forge -->|supports| opp-bio-discovery
    design-savant -->|supports| opp-bio-discovery
    execute-autonomy -->|leads| opp-infra-efficiency
    think-orchestrator -->|supports| opp-infra-efficiency
    identify-scout -->|supports| opp-infra-efficiency
    strategy-supreme -->|supports| opp-infra-efficiency
    design-savant -->|supports| opp-infra-efficiency
    execute-autonomy -->|leads| opp-governance-elevation
    strategy-supreme -->|supports| opp-governance-elevation
    jobRegistry --> opp-supply-inversion
    treasuryManager --> opp-supply-inversion
    a2aBus --> opp-supply-inversion
    worldModel --> opp-latent-alpha
    treasuryManager --> opp-latent-alpha
    governance --> opp-latent-alpha
    jobRegistry --> opp-bio-discovery
    worldModel --> opp-bio-discovery
    a2aBus --> opp-bio-discovery
    treasuryManager --> opp-infra-efficiency
    a2aBus --> opp-infra-efficiency
    governance --> opp-governance-elevation
    a2aBus --> opp-governance-elevation
    metrics[Metrics\nROI 10.77x\nAutomation 92.4%]
    orchestrator --> metrics
All standalone flowcharts (4)

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

18 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-alpha
npm run demo:meta-agentic-alpha:work
npm run demo:meta-agentic-alpha:serve
npm run demo:meta-agentic-alpha:task
npm run demo:meta-agentic-alpha:review
npm run demo:meta-agentic-alpha:worker
npm run demo:meta-agentic-alpha:test
npm run demo:meta-agentic-alpha:qa

Full command catalog and troubleshooting ↗

ORIGINAL DIRECTORYdemo/Meta-Agentic-ALPHA-AGI-Jobs-v0View on GitHub ↗