# idea134-matchops-federated-verifiable

MatchOps is a federated session-orchestration toolkit for multiplayer games. It models regions, servers, session types, capacity budgets, latency budgets, fairness rules, and execution policies in a declarative DSL, then compiles that spec into canonical IR, a verifiable execution graph, and governance artifacts.

This repository focuses on the control-plane primitives:

- declarative DSL compilation
- canonical IR and plan deltas
- cryptographically hashed execution attestations
- a Graph-of-Contracts adapter registry
- a privacy-preserving federated scheduler
- a tamper-evident governance ledger backed by SQLite

## What It Provides

- `MatchOpsSpec`: the input DSL model.
- `compile_spec(...)`: converts a spec into canonical IR and a verifiable execution graph.
- `FederatedOptimizer`: allocates session demand from aggregated regional signals without exposing raw player data.
- `GoCRegistry`: registers versioned adapters for cloud APIs and game-server routing.
- `SQLiteGovernanceLedger`: stores append-only governance events with a hash chain.

## Repository Layout

```text
src/idea134_matchops_federated_verifiable/
tests/
AGENTS.md
test.sh
pyproject.toml
```

## Quick Start

```bash
python3 -m pip install -e .
pytest -q
python3 -m build
```

Or run the full verification script:

```bash
bash test.sh
```

## Design Notes

- Canonical JSON serialization is used for all hashes so decisions are reproducible.
- The execution graph uses per-step attestations derived from the previous step and the current payload.
- The optimizer only consumes aggregated regional signals and session demand, never raw player identity data.
- The governance ledger is append-only and can be persisted to SQLite for durable audit trails.

## Public Package Metadata

The package is published as `idea134-matchops-federated-verifiable` and its README is wired into `pyproject.toml` via `readme = "README.md"`.