feat(relay): rendezvous-relay service — 7 packages + plans (contracts/transport/agent/control-plane/e2e/auth/web)

Multi-tenant reverse-tunnel service ("ngrok for Claude Code" with E2E): a
host-agent dials OUT to an operator-run relay; external devices reach the host
THROUGH the relay, routed by per-tenant subdomain, forwarding ciphertext only
(the relay never sees plaintext). Lets a customer reach their own self-hosted
web-terminal from anywhere with zero networking setup.

Packages — all tsc-strict + vitest green (656 tests), cross-package integration verified:
- relay-contracts: frozen shared contracts (mux frame codec, data model,
  capability token, E2E envelope, pairing) — the src/types.ts analog
- term-relay:   native WS mux + stateless data plane (subdomain routing, ciphertext forward)
- agent:        host-agent (pairing, per-host Ed25519 + mTLS dial-out, forwards to 127.0.0.1:3000)
- control-plane: accounts/hosts registry, pairing-code flow, routing table, provisioning
- relay-e2e:    browser<->agent E2E (X25519 ECDH through relay, AEAD, anti-replay, recoverable replay key)
- relay-auth:   Passkey/WebAuthn, capability tokens, per-host certs, deny-by-default tenant isolation
- relay-web:    browser login + Web Crypto E2E + client-side preview rendering

Security invariants INV1-15 enforced; cross-tenant isolation CI tripwire live
(.github/workflows/relay-tripwire.yml). Design + implementation-level plans in
docs/PLAN_RELAY_*.md and docs/EXPLORE_RELAY_SERVICE.md.

NOTE: generated autonomously per the reviewed plans. The security-critical
packages (relay-e2e, relay-auth) REQUIRE expert security audit before any real
deployment — passing tests prove self-consistency, not resistance to attackers.
Base app (src/, public/) unchanged; concurrent desktop work left uncommitted.

term-relay/test/plugin-hook.test.ts

@@ -0,0 +1,33 @@
+import { describe, test, expect, vi } from 'vitest'
+import { handleFrpHook, type ControlPlaneAuthz } from '../frp-scaffold/plugin-hook.js'
+
+function authz(allow: boolean): ControlPlaneAuthz {
+  return { authorize: vi.fn(async () => ({ allow })) }
+}
+
+const loginReq = { version: '0.1.0', op: 'Login', content: { user: 'alice', token: 't' } }
+
+describe('frp plugin-hook shim (T12, deny-by-default, delegates to P3)', () => {
+  test('unknown token → reject (deny-by-default)', async () => {
+    const res = await handleFrpHook({ ...loginReq, op: 'NewUserConn' }, authz(false))
+    expect(res).toEqual({ reject: true, reject_reason: 'denied by control plane' })
+  })
+
+  test('valid + control-plane-owned subdomain → allow (pass-through unchanged)', async () => {
+    const res = await handleFrpHook(loginReq, authz(true))
+    expect(res).toEqual({ reject: false, unchange: true })
+  })
+
+  test('forwards the decision to P3, never decides tenancy itself', async () => {
+    const a = authz(true)
+    await handleFrpHook({ ...loginReq, op: 'NewProxy' }, a)
+    expect(a.authorize).toHaveBeenCalledWith('NewProxy', loginReq.content)
+  })
+
+  test('malformed payload → reject at the Zod boundary, authorizer NOT called', async () => {
+    const a = authz(true)
+    const res = await handleFrpHook({ op: 'Bogus' }, a)
+    expect(res).toEqual({ reject: true, reject_reason: 'malformed hook request' })
+    expect(a.authorize).not.toHaveBeenCalled()
+  })
+})