feat(rewrite): add Phase 3 monitoring ingestion and catalogs

greenfield/docs/architecture/greenfield-rewrite/application-architecture.md

@@ -370,11 +370,13 @@ outcome. For SSE the response-created event marks
 successful dispatch, not stream termination; close/cancel/error observability remains part of the
 browser/realtime lifecycle slice. Client cancellation is informational and carries neither a
 failure fingerprint nor a server-error outcome. Bun's outer 640 KiB pre-dispatch body ceiling
-supports the largest reviewed task-content request and may reject before application correlation
-exists. The raw tRPC boundary selects exact registered-procedure ceilings before parsing or
-authentication: 16 KiB for authentication, 32 KiB for WebAuthn, 64 KiB by default, 128 KiB for
-task progress, and 640 KiB for task create/content update. Unknown procedures retain the default
-ceiling, while unknown authentication-namespace procedures retain the stricter authentication
+supports the largest reviewed task-content and monitoring mutation requests and may reject before
+application correlation exists. The raw tRPC boundary selects exact registered-procedure ceilings
+before parsing or authentication: 16 KiB for authentication, 32 KiB for WebAuthn, 64 KiB by
+default, 128 KiB for task progress, and 640 KiB for task create/content update, complete monitoring
+snapshots, and
+immutable report upserts. Unknown procedures retain the default ceiling, while unknown
+authentication-namespace procedures retain the stricter authentication
 profile. The raw handler receives the generated ID and resolves direct-client provenance against
 the exact trusted-proxy allowlist before context construction. `createContext` then authenticates the
 already parsed session or automation credential and establishes identity plus audit correlation
@@ -387,7 +389,7 @@ once. Reusable procedure builders are limited to:
 
 Expected errors use a small stable code set such as `UNAUTHORIZED`, `FORBIDDEN`, `CONFLICT`,
 `NOT_FOUND`, `PRECONDITION_FAILED`, `TOO_MANY_REQUESTS`, and `SERVICE_UNAVAILABLE` with safe
-structured details. The `ContractErrorCode` union, all 55 actual router paths, the server-owned
+structured details. The `ContractErrorCode` union, all 68 actual router paths, the server-owned
 runtime allowlist, and generated contract metadata must match exactly. The base procedure
 middleware enforces that allowlist for immediate and deferred subscription failures; an
 implemented procedure missing from the policy or an undeclared code becomes a redacted internal

greenfield/docs/architecture/greenfield-rewrite/progress.md

@@ -12,7 +12,7 @@ closes a phase; dated entries below provide the evidence, not a second status so
 | 0 — Evidence and qualification      | Complete                             | All eight mandatory spikes pass on exact Bun revision `17d6843606d76620cb55d31424d7fb0aed51c367`: build, transport, cross-process SQLite/outbox, Drizzle/Bun SQLite, browser data, chat batching, shutdown, and capped resources. Source-derived parity and the OpenClaw source audit pass as additional evidence.                                                           |
 | 1 — Foundation                      | Complete                             | The self-contained future root builds immutable browser/web/worker artifacts, protects project-local production state, installs exact Bun and systemd artifacts, migrates a database copy, atomically promotes the release/database pair, serves readiness/browser assets, writes project-local logs, and proves crash-safe rollback and shutdown in a disposable lifecycle. |
 | 2 — Trust and transport             | Complete for the stated server scope | Authentication, MFA, WebAuthn, automation credentials, audit, authenticated renewable SSE, one-shot native Gateway bootstrap verification, and the consolidated [threat model](../../security/greenfield-phase-two-threat-model.md) have executable evidence. Browser UI and production cutover remain later gates.                                                          |
-| 3 — Core operator domains           | Started                              | Task and agent-directory parity are implemented with durable history, realtime invalidation, and browser workflows. Report, incident, notification, schedule/job, cache/metrics, overview, and worker-domain parity remain open.                                                                                                                                             |
+| 3 — Core operator domains           | Started                              | Task and agent-directory parity are implemented with durable history, realtime invalidation, and browser workflows. Monitoring ingestion plus report, incident, and notification server parity are implemented. Their browser workflows, schedules/jobs, overview, cache/metrics, and the real worker remain open.                                                           |
 | 4 — Gateway and chat                | Not started                          | The Phase 2 verifier is one-shot only. Persistent native Gateway lifecycle, current-protocol re-audit, sessions, chat journal/recovery, attachments, and frontend remain open.                                                                                                                                                                                               |
 | 5 — Privileged and external domains | Not started                          | Worker-owned file/media, Docker, database, OpenClaw, GitHub, deployment, backup, and other privileged adapters remain open.                                                                                                                                                                                                                                                  |
 | 6 — Parity, hardening, and cutover  | Not started                          | Full UI parity, generated `/docs`, load/resource/restore evidence, cutover rehearsal, fresh production database, and legacy removal remain open.                                                                                                                                                                                                                             |
@@ -841,3 +841,30 @@ full-browser parity, production rehearsal, cutover, and legacy deletion remain o
   Persistent OpenClaw/Gateway availability and session state remain Phase 4 work; reports,
   incidents, notifications, schedules/jobs, overview, cache/metrics, and the real worker remain
   open Phase 3 gates.
+
+### 2026-08-07 — Phase 3 monitoring ingestion and catalog slice
+
+- One automation-only `monitoring.submitCompleteSnapshot` boundary now exposes the complete-run
+  state machine through `monitoring:write`. It shares the production database repository,
+  immediate-write admission, domain clock, and post-commit realtime wake path with the report,
+  incident, and notification catalogs.
+- Twelve catalog procedures provide keyset-paginated incident, notification, and report reads;
+  immutable report upserts for scoped browser sessions or automation callers; automation-only
+  notification producer upserts; session-owned notification read/delete actions; and bounded
+  report deletion. Exact `reports:*` and `notifications:*` capabilities and principal-kind
+  middleware keep browser-session actions separate from automation-only ingestion.
+- Complete snapshots and report upserts have a qualified 640 KiB transport profile plus a stricter
+  512 KiB semantic aggregate budget. Each embedded JSON object retains its separate 64 KiB budget.
+  Exact registered procedure matching prevents unknown, malformed, or batched names from inheriting
+  the larger allowance, while authentication and WebAuthn retain their stricter profiles.
+- Typed validation, replay, catalog conflict, not-found, precondition, and database-admission
+  failures map to the declared tRPC error policy. Mutations clamp durable timestamps across clock
+  regressions, cap report-linked deletion work, and keep catalog writes, audit state, and compact
+  realtime invalidations atomic.
+- Generated procedure and realtime documentation now comes from the same Valibot contracts. The
+  reviewed parity inventory marks all six notification and four report legacy operations
+  implemented; incident reads and complete monitoring ingestion are net-new. Incident-generation
+  notifications retain a forward deep link to the immediately stacked authenticated `/incidents`
+  reader rather than choosing one arbitrary observation report; greenfield stays inactive until
+  the complete cutover stack lands. Browser workflows, schedules/jobs, overview, cache/metrics,
+  and real worker execution remain open Phase 3 gates.

greenfield/docs/generated/procedures.md

@@ -46,6 +46,19 @@
 | `automationSecurity.revokeCredential` | mutation | automation-security | MFA enrollment required; recent MFA when enabled | [input](./schemas/automationSecurity.revokeCredential.input.schema.json) | [output](./schemas/automationSecurity.revokeCredential.output.schema.json) | `CONFLICT`, `FORBIDDEN`, `NOT_FOUND`, `SERVICE_UNAVAILABLE`, `UNAUTHORIZED` | `mfa_enrollment_required`, `step_up_required` | Explicitly revokes one automation credential after client cutover. |
 | `automationSecurity.rotateCredential` | mutation | automation-security | MFA enrollment required; recent MFA when enabled | [input](./schemas/automationSecurity.rotateCredential.input.schema.json) | [output](./schemas/automationSecurity.rotateCredential.output.schema.json) | `CONFLICT`, `FORBIDDEN`, `NOT_FOUND`, `PRECONDITION_FAILED`, `SERVICE_UNAVAILABLE`, `UNAUTHORIZED` | `mfa_enrollment_required`, `step_up_required` | Stages a linked replacement credential without revoking its predecessor. |
 | `events.stream` | subscription | events | Authenticated; per-topic: agents:read, notifications:read, reports:read, tasks:read | [input](./schemas/events.stream.input.schema.json) | [output](./schemas/events.stream.output.schema.json) | `BAD_REQUEST`, `FORBIDDEN`, `SERVICE_UNAVAILABLE`, `TOO_MANY_REQUESTS`, `UNAUTHORIZED` | None | Streams authorized durable changes with tracked resume cursors. |
+| `incidents.get` | query | incidents | Authenticated: reports:read | [input](./schemas/incidents.get.input.schema.json) | [output](./schemas/incidents.get.output.schema.json) | `FORBIDDEN`, `NOT_FOUND`, `UNAUTHORIZED` | None | Loads one exact incident lifecycle record. |
+| `incidents.list` | query | incidents | Authenticated: reports:read | [input](./schemas/incidents.list.input.schema.json) | [output](./schemas/incidents.list.output.schema.json) | `FORBIDDEN`, `UNAUTHORIZED` | None | Lists stable incident lifecycle rows for report navigation. |
+| `monitoring.submitCompleteSnapshot` | mutation | monitoring | Authenticated automation principal: monitoring:write | [input](./schemas/monitoring.submitCompleteSnapshot.input.schema.json) | [output](./schemas/monitoring.submitCompleteSnapshot.output.schema.json) | `BAD_REQUEST`, `CONFLICT`, `FORBIDDEN`, `SERVICE_UNAVAILABLE`, `UNAUTHORIZED` | None | Atomically ingests one complete monitor snapshot. |
+| `notifications.clearRead` | mutation | notifications | Authenticated browser session: notifications:write | [input](./schemas/notifications.clearRead.input.schema.json) | [output](./schemas/notifications.clearRead.output.schema.json) | `FORBIDDEN`, `SERVICE_UNAVAILABLE`, `UNAUTHORIZED` | None | Deletes one bounded page of matching read notifications. |
+| `notifications.delete` | mutation | notifications | Authenticated browser session: notifications:write | [input](./schemas/notifications.delete.input.schema.json) | [output](./schemas/notifications.delete.output.schema.json) | `FORBIDDEN`, `NOT_FOUND`, `SERVICE_UNAVAILABLE`, `UNAUTHORIZED` | None | Deletes one exact Dashboard notification. |
+| `notifications.list` | query | notifications | Authenticated: notifications:read | [input](./schemas/notifications.list.input.schema.json) | [output](./schemas/notifications.list.output.schema.json) | `FORBIDDEN`, `UNAUTHORIZED` | None | Lists a stable filtered notification page and unread count. |
+| `notifications.markAllRead` | mutation | notifications | Authenticated browser session: notifications:write | [input](./schemas/notifications.markAllRead.input.schema.json) | [output](./schemas/notifications.markAllRead.output.schema.json) | `FORBIDDEN`, `SERVICE_UNAVAILABLE`, `UNAUTHORIZED` | None | Marks one bounded page of matching notifications read. |
+| `notifications.markRead` | mutation | notifications | Authenticated browser session: notifications:write | [input](./schemas/notifications.markRead.input.schema.json) | [output](./schemas/notifications.markRead.output.schema.json) | `FORBIDDEN`, `NOT_FOUND`, `SERVICE_UNAVAILABLE`, `UNAUTHORIZED` | None | Marks one notification read idempotently. |
+| `notifications.upsert` | mutation | notifications | Authenticated automation principal: notifications:write | [input](./schemas/notifications.upsert.input.schema.json) | [output](./schemas/notifications.upsert.output.schema.json) | `BAD_REQUEST`, `CONFLICT`, `FORBIDDEN`, `NOT_FOUND`, `SERVICE_UNAVAILABLE`, `UNAUTHORIZED` | None | Creates a notification or accepts an exact idempotent replay. |
+| `reports.delete` | mutation | reports | Authenticated: reports:write | [input](./schemas/reports.delete.input.schema.json) | [output](./schemas/reports.delete.output.schema.json) | `FORBIDDEN`, `NOT_FOUND`, `PRECONDITION_FAILED`, `SERVICE_UNAVAILABLE`, `UNAUTHORIZED` | None | Deletes one report while preserving its monitor-run history. |
+| `reports.get` | query | reports | Authenticated: reports:read | [input](./schemas/reports.get.input.schema.json) | [output](./schemas/reports.get.output.schema.json) | `FORBIDDEN`, `NOT_FOUND`, `UNAUTHORIZED` | None | Loads one complete immutable Markdown report. |
+| `reports.list` | query | reports | Authenticated: reports:read | [input](./schemas/reports.list.input.schema.json) | [output](./schemas/reports.list.output.schema.json) | `FORBIDDEN`, `UNAUTHORIZED` | None | Lists a stable filtered page of immutable report summaries. |
+| `reports.upsert` | mutation | reports | Authenticated: reports:write | [input](./schemas/reports.upsert.input.schema.json) | [output](./schemas/reports.upsert.output.schema.json) | `BAD_REQUEST`, `CONFLICT`, `FORBIDDEN`, `SERVICE_UNAVAILABLE`, `UNAUTHORIZED` | None | Creates a report or accepts an exact idempotent replay. |
 | `securityAudit.listEvents` | query | securityAudit | Authenticated browser session | [input](./schemas/securityAudit.listEvents.input.schema.json) | [output](./schemas/securityAudit.listEvents.output.schema.json) | `FORBIDDEN`, `UNAUTHORIZED` | None | Lists redacted immutable security events in stable newest-first order. |
 | `system.runtimeIdentity` | query | system | Public | [input](./schemas/system.runtimeIdentity.input.schema.json) | [output](./schemas/system.runtimeIdentity.output.schema.json) | None | None | Returns the Bun runtime identity of the serving process. |
 | `tasks.addUpdate` | mutation | tasks | Authenticated: tasks:write | [input](./schemas/tasks.addUpdate.input.schema.json) | [output](./schemas/tasks.addUpdate.output.schema.json) | `CONFLICT`, `FORBIDDEN`, `NOT_FOUND`, `SERVICE_UNAVAILABLE`, `UNAUTHORIZED` | None | Appends one authenticated progress update to a task. |

greenfield/docs/generated/realtime-events.md

@@ -2,4 +2,10 @@
 
 > Generated by `bun run docs:generate`. Do not edit manually.
 
-No standalone realtime topic references are published until their snapshot procedures are implemented.
+| Topic | Payload | Snapshot | Retention | Summary |
+| --- | --- | --- | --- | --- |
+| `agents.status` | [payload](./schemas/agents.status.realtime.payload.schema.json) | `agents.listStatuses` | 7 days | Invalidates one agent status row after a durable metadata change. |
+| `monitoring.incidents` | [payload](./schemas/monitoring.incidents.realtime.payload.schema.json) | `incidents.list` | 7 days | Invalidates incident lifecycle rows after a complete monitor snapshot. |
+| `monitoring.notifications` | [payload](./schemas/monitoring.notifications.realtime.payload.schema.json) | `notifications.list` | 7 days | Invalidates Dashboard notifications after catalog changes. |
+| `monitoring.reports` | [payload](./schemas/monitoring.reports.realtime.payload.schema.json) | `reports.list` | 7 days | Invalidates immutable reports after catalog changes. |
+| `tasks.records` | [payload](./schemas/tasks.records.realtime.payload.schema.json) | `tasks.list` | 7 days | Invalidates one task row after a durable task-domain change. |

greenfield/docs/generated/schemas/agents.status.realtime.payload.schema.json

@@ -0,0 +1,17 @@
+{
+  "$id": "urn:mira-dashboard:agents.status.realtime.payload",
+  "type": "object",
+  "properties": {
+    "id": {
+      "type": "string",
+      "minLength": 1,
+      "maxLength": 64,
+      "pattern": "^[a-z0-9][a-z0-9._-]*$"
+    }
+  },
+  "required": [
+    "id"
+  ],
+  "additionalProperties": false,
+  "$schema": "https://json-schema.org/draft/2020-12/schema"
+}

greenfield/docs/generated/schemas/automationSecurity.createPrincipal.input.schema.json

@@ -8,14 +8,17 @@
         "enum": [
           "agents:read",
           "agents:write",
+          "monitoring:write",
           "notifications:read",
+          "notifications:write",
           "reports:read",
+          "reports:write",
           "tasks:read",
           "tasks:write"
         ],
         "type": "string"
       },
-      "maxItems": 6,
+      "maxItems": 9,
       "uniqueItems": true
     },
     "id": {

greenfield/docs/generated/schemas/automationSecurity.createPrincipal.output.schema.json

@@ -77,14 +77,17 @@
                 "enum": [
                   "agents:read",
                   "agents:write",
+                  "monitoring:write",
                   "notifications:read",
+                  "notifications:write",
                   "reports:read",
+                  "reports:write",
                   "tasks:read",
                   "tasks:write"
                 ],
                 "type": "string"
               },
-              "maxItems": 6,
+              "maxItems": 9,
               "uniqueItems": true
             },
             "createdAtMs": {
@@ -150,14 +153,17 @@
                 "enum": [
                   "agents:read",
                   "agents:write",
+                  "monitoring:write",
                   "notifications:read",
+                  "notifications:write",
                   "reports:read",
+                  "reports:write",
                   "tasks:read",
                   "tasks:write"
                 ],
                 "type": "string"
               },
-              "maxItems": 6,
+              "maxItems": 9,
               "uniqueItems": true
             },
             "createdAtMs": {

greenfield/docs/generated/schemas/automationSecurity.disablePrincipal.output.schema.json

@@ -26,14 +26,17 @@
                 "enum": [
                   "agents:read",
                   "agents:write",
+                  "monitoring:write",
                   "notifications:read",
+                  "notifications:write",
                   "reports:read",
+                  "reports:write",
                   "tasks:read",
                   "tasks:write"
                 ],
                 "type": "string"
               },
-              "maxItems": 6,
+              "maxItems": 9,
               "uniqueItems": true
             },
             "createdAtMs": {
@@ -99,14 +102,17 @@
                 "enum": [
                   "agents:read",
                   "agents:write",
+                  "monitoring:write",
                   "notifications:read",
+                  "notifications:write",
                   "reports:read",
+                  "reports:write",
                   "tasks:read",
                   "tasks:write"
                 ],
                 "type": "string"
               },
-              "maxItems": 6,
+              "maxItems": 9,
               "uniqueItems": true
             },
             "createdAtMs": {