Read the launch playbook
Inspect the operational journey and its assumptions. The original checklist and owner field guide remain available in the complete library below.
demo/astral-omnidominion-operating-system-command-theatre/launch-playbook.md ↗Governance & validation
Explore the command-theatre presentation of Astral coordination and control.
A CLOSER LOOK
For presenters and operators: turn the command-theatre documents into a local coverage report and explain exactly how its heuristic scores are produced.
Inspect the operational journey and its assumptions. The original checklist and owner field guide remain available in the complete library below.
demo/astral-omnidominion-operating-system-command-theatre/launch-playbook.md ↗Coverage counts available documents; the coordination, Gibbs and payoff scores are functions of document count and line count. They are documentation heuristics, not physical or economic measurements.
demo/astral-omnidominion-operating-system-command-theatre/run_demo.py ↗Use the checklist to organize an actual review. A high documentation score cannot establish that any checklist action has been completed.
demo/astral-omnidominion-operating-system-command-theatre/mission-review-checklist.md ↗REAL REPOSITORY MATERIAL
Reading the exact source at revision 5b4cebb3.
This browser inspection does not execute the demo.
demo/astral-omnidominion-operating-system-command-theatre/launch-playbook.md
Select a walkthrough step to explore its source.
# Launch Playbook — Astral Omnidominion Command Theatre
Begin with the offline rehearsal. Use the full AGI OS stack when you are ready to test the separate services and contracts.
## 1. Offline rehearsal (default)
Requirement: Python 3.10 or later (`python3 --version`). From the repository root:
```bash
python3 demo/astral-omnidominion-operating-system-command-theatre/run_demo.py
```
Open `reports/astral-omnidominion-operating-system-command-theatre/report.html`. The page works offline, adapts to narrow screens and provides file links, readable outcomes and a ten-job catalog. There are no external scripts, fonts or CDN dependencies.
Expected: `accepted`, six checks passed, three unique ledger entries, one duplicate, 21,999 cents total. Read `report.json` for machine-readable fields. Each execution stores candidate files in a new `runs/<id>/` directory; it does not reuse another run's artifacts. Each completed task also retains its own `receipt.json` and `index.html` inside that directory, so earlier runs stay independently verifiable. The top-level report/dashboard represent the latest invocation at that output path.
- `--scenario rejected`: an intentional one-cent error is detected; exit 1.
- `--scenario paused`: no task is executed and no task artifacts are written; exit 1.
- `--output /absolute/path/report.json`: write a separate report and adjacent dashboard. Output must end in `.json` and stay outside the demo source directory.
- `--verify-report /absolute/path/report.json`: verify all artifact hashes and recompute acceptance; does not execute work or contact services.
Use different output paths for demonstrations you want to compare. The current report and its referenced `runs/` directory form one evidence package. A local hash is not a signed attestation.
## 2. Full AGI OS stack (advanced)
This existing path may deploy contracts and start services. It is distinct from the offline rehearsal and the OpenClaw worker. Use disposable local/testnet environments and inspect configured RPCs and signers before execution.
| Requirement | Verification |
| --- | --- |
| Repository-pinned Node and npm | Follow [START_HERE](../../docs/START_HERE.md); `.nvmrc` and `package.json` are authoritative |
| Locked dependencies | `npm ci` from the repository root |
| Docker with Compose, or a prepared compatible runtime | `docker --version`, `docker compose version`, `docker compose ps` |
| Local chain and deployment configuration | Review [the parent OS demo](../astral-omnidominion-operating-system/README.md) and `deployment-config/` |
| Git and sufficient storage | `git status`, `df -h .`; allow space for images and artifacts |
For an explicit local deployment with Compose:
```bash
npm run demo:agi-os:first-class -- --network localhost --compose
```
For a previously prepared local deployment:
```bash
npm run demo:agi-os:first-class -- --network localhost --yes --no-compose --skip-deploy
```
The script supports `localhost` and `sepolia`. “Local Hardhat (Anvil)” is its existing preset label; Hardhat and Anvil are different tools. Check which runtime is actually running. Do not select a funded network for the first rehearsal.
The orchestrator runs preflight, the deployment wizard when enabled, `demo:agi-os`, owner diagram generation, owner-control verification, summary rendering, manifest compilation and integrity checks. **Steps can be skipped when dependencies or artifacts are unavailable, even when the command exits zero.** Inspect `first-class-run.json` → `steps[]`; a skipped check is not a pass. `--skip-deploy` only skips the deployment wizard, not every side effect of later stages. These commands do not run the entire repository test/security suite.
### Live mission control
Inspect actual port bindings and enabled services with `docker compose ps` before opening a UI. The Compose file maps `validator-ui` to `http://localhost:3000` and enterprise portal to `http://localhost:3001`. A validator UI is not guaranteed at port 3002. The separate static console can be built and previewed with:
```bash
npm run webapp:build
npm --prefix apps/console run preview -- --host 127.0.0.1 --port 4173
```
Confirm the actual network, contract addresses, account and data mode before submitting a task or governance action. Inspect transactions and refreshed authoritative state; a green button, static demo mode or polling animation is not proof of settlement.
### Full-stack mission evidence
Under `reports/agi-os/`, retain:
- `grand-summary.md` / `.html` — generated overview.
- `owner-control-matrix.json` — module surfaces and configuration status.
- `mission-bundle/` — simulation logs, telemetry and manifest when generated.
- `first-class/first-class-run.json` — host/commit metadata and each stage's status.
- `first-class/first-class-manifest.json` — SHA-256 entries, not signed attestations.
- `first-class/logs/` — command logs.
- `first-class/owner-control-map.mmd` — existing Mermaid governance map when generated.
Apply the [mission review checklist](mission-review-checklist.md). Do not reuse stale outputs to claim a newly skipped step succeeded.
## 3. Shutdown and troubleshooting
To stop the local Compose stack without deleting named volumes:
```bash
docker compose -f compose.yaml down
```
`docker compose -f compose.yaml down -v` additionally deletes named volumes. Use that only for an intentionally disposable environment after retaining needed evidence. Deployment reruns can change state; they are not guaranteed idempotent.
| Symptom | Next step |
| --- | --- |
| `python3` missing or older than 3.10 | Install a supported Python; the offline path needs no pip packages |
| Missing/empty document or invalid catalog | Restore the named source file; rerun and inspect the new report |
| Intentional rejected/paused scenario exits 1 | Expected; inspect the status and failed/blocked condition |
| Integrity check fails | Keep the evidence, identify changed/missing files and generate a new run; do not hand-edit hashes to make an old result pass |
| Output permission error | Choose a writable report directory outside the demo sources |
| Full-stack stage skipped | Satisfy that stage's prerequisites and rerun; zero exit alone is insufficient |
| Provider task interrupted | Stop and reconcile using the [computer-work recovery guide](../../docs/computer-work.md#recovery-and-stop-controls) |
SHA-256 85650bf307ce3cd6e90303b442a0d1212b4c27f027534ebbf2ed4c68bc482602
FROM READING TO A REPRODUCIBLE RUN
Run from the repository root with Python 3.12. This selected entry point uses the Python standard library. Keep output in a separate directory so that you can compare runs.
Complete environment setup ↗python demo/astral-omnidominion-operating-system-command-theatre/run_demo.py --output /tmp/astral-theatre.jsonA local JSON report records document coverage and bounded heuristic scores. Inspect each document entry and the formulas that produce the scores.
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
Compare the coverage result with the actual document list. Explain why adding prose could increase a score without improving operational readiness.
THE SYSTEM, MADE VISIBLE
flowchart LR
Operators((Mission Owners)) --> demo_astral_omnidominion_operating_system_command_theatre[[Demo → Astral Omnidominion Operating System Command Theatre]]
demo_astral_omnidominion_operating_system_command_theatre --> Core[[AGI Jobs v0 (v2) Core Intelligence]]
Core --> Observability[[Unified CI / CD & Observability]]
Core --> Governance[[Owner Control Plane]]WHEN SOMETHING DOESN’T MATCH
Confirm the active virtual environment and the selected demo’s requirements. Run python -m pip check; do not install unrelated demo requirements over a working environment.
Check the selected entry point, configuration and output argument. Keep the seed and implementation fixed before comparing outcomes.
TRACE THE CHECKS
2 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.
For live commissioning, consult the production readiness record.
EVERY VARIANT, PRESERVED
REPRODUCE & INSPECT
Run commands from the repository root after following this demo's guide. Network and owner actions require their documented setup.
No root-level launch command is associated with this source path. Follow the guide or source directory for its own entry point.