feat: ground zero-touch forge readiness (#912)

config/zero_touch_forge.json

@@ -0,0 +1,54 @@
+{
+  "epic_issue": 912,
+  "title": "The Zero-Touch Forge: Bare-Metal Fleet Bootstrap in 60 Minutes",
+  "checks": [
+    {
+      "id": "os_bootstrap",
+      "label": "OS bootstrap foothold",
+      "required_files": ["scripts/provision-runner.sh"],
+      "required_signals": []
+    },
+    {
+      "id": "integrity_validation",
+      "label": "Repository integrity validation",
+      "required_files": [],
+      "required_signals": ["has_crypto_integrity_verification"]
+    },
+    {
+      "id": "secret_distribution",
+      "label": "Encrypted seed / secret distribution",
+      "required_files": [],
+      "required_signals": ["has_age_seed_flow"]
+    },
+    {
+      "id": "stack_startup",
+      "label": "Full stack startup manifest",
+      "required_files": ["docker-compose.yml", "fleet/fleet-routing.json"],
+      "required_signals": ["has_stack_start_manifest"]
+    },
+    {
+      "id": "test_gate",
+      "label": "Bootstrap test gate",
+      "required_files": [],
+      "required_signals": ["has_test_gate"]
+    },
+    {
+      "id": "checkpoint_restore",
+      "label": "Checkpoint restore primitive",
+      "required_files": ["scripts/lazarus_checkpoint.py"],
+      "required_signals": []
+    },
+    {
+      "id": "post_boot_notification",
+      "label": "Post-boot notify Alexander only-after-healthy",
+      "required_files": [],
+      "required_signals": ["has_notification_step"]
+    },
+    {
+      "id": "sixty_minute_sla",
+      "label": "60-minute end-to-end timing budget",
+      "required_files": [],
+      "required_signals": ["has_sla_budget"]
+    }
+  ]
+}

docs/zero-touch-forge-readiness.md

@@ -0,0 +1,51 @@
+# Zero-Touch Forge Readiness
+
+Epic: #912 — The Zero-Touch Forge: Bare-Metal Fleet Bootstrap in 60 Minutes
+
+## Impossible Goal
+
+Take a raw VPS plus only a git URL and encrypted seed, then bootstrap a full Timmy Foundation fleet in under 60 minutes with no human intervention after trigger.
+
+This document does **not** claim the goal is solved. It grounds the epic in the current repo state.
+
+Current primitive readiness: 2 ready / 6 blocked.
+
+## Current Readiness Table
+
+| Check | Status | Evidence | Missing Pieces |
+|-------|--------|----------|----------------|
+| OS bootstrap foothold | READY | scripts/provision-runner.sh=present | — |
+| Repository integrity validation | BLOCKED | has_crypto_integrity_verification=no | has_crypto_integrity_verification |
+| Encrypted seed / secret distribution | BLOCKED | has_age_seed_flow=no | has_age_seed_flow |
+| Full stack startup manifest | BLOCKED | docker-compose.yml=present, fleet/fleet-routing.json=present, has_stack_start_manifest=no | has_stack_start_manifest |
+| Bootstrap test gate | BLOCKED | has_test_gate=no | has_test_gate |
+| Checkpoint restore primitive | READY | scripts/lazarus_checkpoint.py=present | — |
+| Post-boot notify Alexander only-after-healthy | BLOCKED | has_notification_step=no | has_notification_step |
+| 60-minute end-to-end timing budget | BLOCKED | has_sla_budget=no | has_sla_budget |
+
+## Interpretation
+
+### What already exists
+- `scripts/provision-runner.sh` proves we already automate part of bare-metal service bootstrap.
+- `scripts/lazarus_checkpoint.py` proves we already have a checkpoint / restore primitive for mission state.
+- `docker-compose.yml`, `fleet/fleet-routing.json`, `operations/fleet-topology.md`, and `config/fleet_agents.json` show a real fleet shape, not just a philosophical wish.
+
+### What is still missing
+- no verified cryptographic repo-integrity gate for a cold bootstrap run
+- no age-encrypted seed / recovery-bundle path in this repo
+- no single stack-start manifest that can bring up Gitea, Nostr relay, Ollama, and all agents from bare metal
+- no bootstrap test gate that refuses health until the full stack passes
+- no explicit notify-Alexander-only-after-healthy step
+- no measured 60-minute execution budget proving the impossible bar
+
+## Next Concrete Build Steps
+
+1. Add an age-based recovery bundle flow and a decrypt/distribute bootstrap primitive.
+2. Add a single stack-start manifest that covers Gitea + relay + Ollama + agent services from one command.
+3. Add a zero-touch health gate script that verifies the full stack before declaring success.
+4. Add a post-boot notification step that only fires after the health gate is green.
+5. Add a timed rehearsal harness so the 60-minute claim can be measured instead of imagined.
+
+## Honest Bottom Line
+
+The repo already contains useful bootstrap and recovery primitives, but it does **not** yet implement a true zero-touch forge. The epic remains open because the hard problems — trust bootstrapping, full-stack orchestration, and timed self-verification — are still unresolved.

scripts/zero_touch_forge_readiness.py

@@ -0,0 +1,187 @@
+#!/usr/bin/env python3
+"""Zero-Touch Forge readiness grounding for epic #912.
+
+This does not pretend the impossible goal is solved.
+It computes which primitive building blocks already exist in the repo and which
+critical gaps still block a true zero-touch forge.
+"""
+
+from __future__ import annotations
+
+import argparse
+import json
+from pathlib import Path
+from typing import Any
+
+REPO_ROOT = Path(__file__).resolve().parent.parent
+SPEC_PATH = REPO_ROOT / "config" / "zero_touch_forge.json"
+
+
+def load_spec(path: Path | None = None) -> dict[str, Any]:
+    target = path or SPEC_PATH
+    return json.loads(target.read_text())
+
+
+def _file_exists_map(repo_root: Path, paths: list[str]) -> dict[str, bool]:
+    return {path: (repo_root / path).exists() for path in paths}
+
+
+def _agent_count(repo_root: Path) -> int:
+    config_path = repo_root / "config" / "fleet_agents.json"
+    if not config_path.exists():
+        return 0
+    try:
+        payload = json.loads(config_path.read_text())
+        return len(payload.get("agents") or [])
+    except Exception:
+        return 0
+
+
+def derive_signal_flags(repo_root: Path | None = None) -> dict[str, bool]:
+    root = repo_root or REPO_ROOT
+    agent_count = _agent_count(root)
+    return {
+        "has_age_seed_flow": False,
+        "has_crypto_integrity_verification": False,
+        "has_stack_start_manifest": agent_count >= 5,
+        "has_test_gate": False,
+        "has_notification_step": False,
+        "has_sla_budget": False,
+    }
+
+
+def _evidence_line(check: dict[str, Any], file_exists: dict[str, bool], signal_flags: dict[str, bool]) -> str:
+    parts = []
+    for path in check.get("required_files", []):
+        parts.append(f"{path}={'present' if file_exists.get(path) else 'missing'}")
+    for key in check.get("required_signals", []):
+        parts.append(f"{key}={'yes' if signal_flags.get(key) else 'no'}")
+    return ", ".join(parts) if parts else "no explicit evidence"
+
+
+def evaluate_readiness(
+    spec: dict[str, Any],
+    *,
+    file_exists: dict[str, bool] | None = None,
+    signal_flags: dict[str, bool] | None = None,
+) -> dict[str, Any]:
+    all_paths = []
+    for check in spec["checks"]:
+        all_paths.extend(check.get("required_files", []))
+
+    file_exists = file_exists or _file_exists_map(REPO_ROOT, sorted(set(all_paths)))
+    signal_flags = signal_flags or derive_signal_flags(REPO_ROOT)
+
+    ready_checks = []
+    blocked_checks = []
+    checks = []
+
+    for check in spec["checks"]:
+        missing_files = [path for path in check.get("required_files", []) if not file_exists.get(path, False)]
+        missing_signals = [key for key in check.get("required_signals", []) if not signal_flags.get(key, False)]
+        ready = not missing_files and not missing_signals
+        result = {
+            "id": check["id"],
+            "label": check["label"],
+            "ready": ready,
+            "missing_files": missing_files,
+            "missing_signals": missing_signals,
+            "evidence": _evidence_line(check, file_exists, signal_flags),
+        }
+        checks.append(result)
+        if ready:
+            ready_checks.append(result)
+        else:
+            blocked_checks.append(result)
+
+    return {
+        "epic_issue": spec["epic_issue"],
+        "title": spec["title"],
+        "ready_count": len(ready_checks),
+        "blocked_count": len(blocked_checks),
+        "ready_checks": ready_checks,
+        "blocked_checks": blocked_checks,
+        "checks": checks,
+        "signals": signal_flags,
+        "files": file_exists,
+    }
+
+
+def render_markdown(report: dict[str, Any]) -> str:
+    lines = [
+        "# Zero-Touch Forge Readiness",
+        "",
+        f"Epic: #{report['epic_issue']} — {report['title']}",
+        "",
+        "## Impossible Goal",
+        "",
+        "Take a raw VPS plus only a git URL and encrypted seed, then bootstrap a full Timmy Foundation fleet in under 60 minutes with no human intervention after trigger.",
+        "",
+        "This document does **not** claim the goal is solved. It grounds the epic in the current repo state.",
+        "",
+        f"Current primitive readiness: {report['ready_count']} ready / {report['blocked_count']} blocked.",
+        "",
+        "## Current Readiness Table",
+        "",
+        "| Check | Status | Evidence | Missing Pieces |",
+        "|-------|--------|----------|----------------|",
+    ]
+    for check in report["checks"]:
+        status = "READY" if check["ready"] else "BLOCKED"
+        missing = ", ".join(check["missing_files"] + check["missing_signals"]) or "—"
+        lines.append(f"| {check['label']} | {status} | {check['evidence']} | {missing} |")
+
+    lines.extend([
+        "",
+        "## Interpretation",
+        "",
+        "### What already exists",
+        "- `scripts/provision-runner.sh` proves we already automate part of bare-metal service bootstrap.",
+        "- `scripts/lazarus_checkpoint.py` proves we already have a checkpoint / restore primitive for mission state.",
+        "- `docker-compose.yml`, `fleet/fleet-routing.json`, `operations/fleet-topology.md`, and `config/fleet_agents.json` show a real fleet shape, not just a philosophical wish.",
+        "",
+        "### What is still missing",
+        "- no verified cryptographic repo-integrity gate for a cold bootstrap run",
+        "- no age-encrypted seed / recovery-bundle path in this repo",
+        "- no single stack-start manifest that can bring up Gitea, Nostr relay, Ollama, and all agents from bare metal",
+        "- no bootstrap test gate that refuses health until the full stack passes",
+        "- no explicit notify-Alexander-only-after-healthy step",
+        "- no measured 60-minute execution budget proving the impossible bar",
+        "",
+        "## Next Concrete Build Steps",
+        "",
+        "1. Add an age-based recovery bundle flow and a decrypt/distribute bootstrap primitive.",
+        "2. Add a single stack-start manifest that covers Gitea + relay + Ollama + agent services from one command.",
+        "3. Add a zero-touch health gate script that verifies the full stack before declaring success.",
+        "4. Add a post-boot notification step that only fires after the health gate is green.",
+        "5. Add a timed rehearsal harness so the 60-minute claim can be measured instead of imagined.",
+        "",
+        "## Honest Bottom Line",
+        "",
+        "The repo already contains useful bootstrap and recovery primitives, but it does **not** yet implement a true zero-touch forge. The epic remains open because the hard problems — trust bootstrapping, full-stack orchestration, and timed self-verification — are still unresolved.",
+        "",
+    ])
+    return "\n".join(lines)
+
+
+def parse_args() -> argparse.Namespace:
+    parser = argparse.ArgumentParser(description="Evaluate repo readiness for the Zero-Touch Forge epic.")
+    parser.add_argument("--json", action="store_true", help="Emit JSON instead of markdown")
+    parser.add_argument("--out", type=Path, help="Optional output file")
+    return parser.parse_args()
+
+
+def main() -> None:
+    args = parse_args()
+    spec = load_spec()
+    report = evaluate_readiness(spec)
+    output = json.dumps(report, indent=2) if args.json else render_markdown(report)
+    if args.out:
+        args.out.parent.mkdir(parents=True, exist_ok=True)
+        args.out.write_text(output)
+    else:
+        print(output)
+
+
+if __name__ == "__main__":
+    main()

tests/test_zero_touch_forge_readiness.py

@@ -0,0 +1,67 @@
+from pathlib import Path
+import sys
+
+sys.path.insert(0, str(Path(__file__).resolve().parent.parent))
+
+from scripts.zero_touch_forge_readiness import evaluate_readiness, load_spec
+
+
+DOC = Path("docs/zero-touch-forge-readiness.md")
+
+
+def test_load_spec_contains_all_impossible_bar_checks():
+    spec = load_spec()
+    check_ids = [item["id"] for item in spec["checks"]]
+    assert check_ids == [
+        "os_bootstrap",
+        "integrity_validation",
+        "secret_distribution",
+        "stack_startup",
+        "test_gate",
+        "checkpoint_restore",
+        "post_boot_notification",
+        "sixty_minute_sla",
+    ]
+
+
+def test_evaluate_readiness_marks_missing_components_as_blockers():
+    spec = load_spec()
+    result = evaluate_readiness(
+        spec,
+        file_exists={
+            "scripts/provision-runner.sh": True,
+            "scripts/lazarus_checkpoint.py": True,
+            "operations/fleet-topology.md": True,
+            "docker-compose.yml": False,
+            "fleet/fleet-routing.json": False,
+            "tests/test_bootstrap_contract.py": False,
+        },
+        signal_flags={
+            "has_age_seed_flow": False,
+            "has_crypto_integrity_verification": False,
+            "has_stack_start_manifest": False,
+            "has_test_gate": False,
+            "has_notification_step": False,
+            "has_sla_budget": False,
+        },
+    )
+
+    assert result["ready_count"] == 2
+    blocked = {item["id"] for item in result["blocked_checks"]}
+    assert blocked == {
+        "integrity_validation",
+        "secret_distribution",
+        "stack_startup",
+        "test_gate",
+        "post_boot_notification",
+        "sixty_minute_sla",
+    }
+
+
+def test_document_exists_with_required_sections():
+    assert DOC.exists(), "expected zero-touch forge readiness doc to exist"
+    content = DOC.read_text()
+    assert "# Zero-Touch Forge Readiness" in content
+    assert "## Impossible Goal" in content
+    assert "## Current Readiness Table" in content
+    assert "## Next Concrete Build Steps" in content