fix(desktop): keep OS clipboard chords (Cmd+C/V/X/A) bubbling out of v1 terminal

apps/desktop/src/renderer/screens/main/components/WorkspaceView/ContentView/TabsContent/Terminal/helpers.ts

@@ -586,6 +586,50 @@ export function setupKeyboardHandler(
 			return false;
 		}
 
+		// OS-level clipboard chords must always bubble so the host (Electron /
+		// xterm.js Selection / OS) can fulfil copy / paste / cut / select-all.
+		// Without this, #3390's "unregistered → PTY" rule swallows Cmd+C/V/X/A
+		// (Mac) and the Linux/Windows clipboard chords, breaking text copy
+		// from the terminal pane. Regression of #3390.
+		const onlyMeta =
+			event.metaKey && !event.ctrlKey && !event.altKey && !event.shiftKey;
+		const ctrlShiftOnly =
+			event.ctrlKey && event.shiftKey && !event.metaKey && !event.altKey;
+		const onlyCtrl =
+			event.ctrlKey && !event.metaKey && !event.altKey && !event.shiftKey;
+		const onlyShift =
+			event.shiftKey && !event.ctrlKey && !event.metaKey && !event.altKey;
+		if (isMac && onlyMeta) {
+			// Cmd+C copy, Cmd+V paste, Cmd+X cut, Cmd+A select-all
+			if (
+				event.code === "KeyC" ||
+				event.code === "KeyV" ||
+				event.code === "KeyX" ||
+				event.code === "KeyA"
+			) {
+				return false;
+			}
+		}
+		if (!isMac) {
+			// Ctrl+Shift+C/V/X/A — Linux/Windows terminal clipboard chords.
+			if (
+				ctrlShiftOnly &&
+				(event.code === "KeyC" ||
+					event.code === "KeyV" ||
+					event.code === "KeyX" ||
+					event.code === "KeyA")
+			) {
+				return false;
+			}
+			// Ctrl+Insert = copy, Shift+Insert = paste. Deliberately exclude
+			// Ctrl+Delete (delete-word-forward in shells) and Shift+Delete
+			// (ambiguous; many terminals treat as readline edit) — those must
+			// reach the PTY.
+			if (event.code === "Insert" && (onlyCtrl || onlyShift)) {
+				return false;
+			}
+		}
+
 		// Only bubble chords registered as app hotkeys; everything else reaches the PTY.
 		// Mirrors v2 terminal-runtime.ts:21 (VSCode terminalInstance pattern).
 		if (resolveHotkeyFromEvent(event) !== null) return false;

docs/observability/event-taxonomy.md

@@ -0,0 +1,11 @@
+# Event Taxonomy
+
+Blueprint refs: P1, P2, P6, P8, P19.
+
+Status: seed. Add events only through typed helpers/proxy; do not scatter vendor SDK calls.
+
+| Event | Journey | When | Required fields | Forbidden fields |
+| --- | --- | --- | --- | --- |
+| `superset.journey_started` | TBD | User/system intent accepted | `correlation_id`, `surface`, `environment`, `app_version` | raw content, token, email, phone |
+| `superset.journey_completed` | TBD | Business success confirmed | `correlation_id`, `result`, `success` | raw content, token, email, phone |
+| `superset.journey_failed` | TBD | Bounded failure class known | `correlation_id`, `error_class`, `success` | raw error message as label |

docs/observability/field-registry.md

@@ -0,0 +1,25 @@
+# Observability Field Registry
+
+Blueprint source: company-wide `observability-blueprint.md`. This repo adapts the contract; it does not fork the blueprint.
+
+| Field | Type | Headers | Logs | Spans | Metric labels | Notes |
+| --- | --- | --- | --- | --- | --- | --- |
+| `request_id` | generated id | yes | yes | yes | no | One HTTP/IPC/job operation. |
+| `trace_id` | W3C trace id | `traceparent` | yes | native | no | Full journey trace. |
+| `span_id` | W3C span id | `traceparent` | yes | native | no | Current operation span. |
+| `correlation_id` | generated id | yes | yes | yes | no | One business action/retry chain. |
+| `actor_id` | hashed/pseudonymous id | yes | yes | yes | no | Never raw email/phone/customer id. |
+| `session_id` | session id | optional | yes | yes | no | Browser/app session. |
+| `app_version` | bounded string | yes | yes | yes | yes | Release/git SHA/build id. |
+| `environment` | enum | optional | yes | yes | yes | `development`, `staging`, `production`. |
+| `surface` | enum | yes | yes | yes | yes | `web`, `api`, `worker`, `job`, `desktop`, `mobile`, `gateway`. |
+| `route` | template | no | yes | yes | yes | No raw URLs or query strings. |
+| `operation` | bounded string | optional | yes | yes | yes | Business operation name. |
+| `result` | enum | optional | yes | yes | yes | `success`, `failure`, `cancelled`, `timeout`, `fallback`. |
+| `success` | string enum | optional | yes | yes | yes | Use `"true"`/`"false"` for labels. |
+| `error_class` | bounded enum | optional | yes | yes | yes | Never raw error message as a label. |
+
+Rules:
+- IDs are fine in logs/traces, not metric labels.
+- Raw stable identifiers stay local unless explicitly approved.
+- Telemetry helpers must redact tokens, cookies, emails, phones, prompts, raw files, webhook URLs, and URL query secrets.

docs/observability/journeys.md

@@ -0,0 +1,23 @@
+# Observability Journeys
+
+## J1 — user sends chat/tool request
+
+Start: TBD.
+Success: TBD.
+Failure/timeout/cancel: TBD.
+Required signals: product event, structured log, metric, trace/span.
+
+## J2 — runtime executes provider call
+
+Start: TBD.
+Success: TBD.
+Failure/timeout/cancel: TBD.
+Required signals: product event, structured log, metric, trace/span.
+
+## J3 — desktop/web session recovers from error
+
+Start: TBD.
+Success: TBD.
+Failure/timeout/cancel: TBD.
+Required signals: product event, structured log, metric, trace/span.
+

docs/observability/query-reference.md

@@ -0,0 +1,34 @@
+# Observability Query Reference
+
+Status: initial repo contract. Fill datasource names when Grafana/OTel/PostHog/Sentry wiring lands.
+
+## Canonical labels
+
+- `environment`: `development`, `staging`, `production`
+- `surface`: `web`, `api`, `worker`, `job`, `desktop`, `mobile`, `gateway`
+- `success`: `"true"` or `"false"`
+- `route`: route templates only
+- `operation`: bounded business operation names
+
+## Baseline metrics expected
+
+- `api_requests_total` or repo equivalent
+- `http_server_request_duration_seconds`
+- `job_runs_total`
+- `job_duration_seconds`
+- `external_dependency_requests_total`
+- `external_dependency_duration_seconds`
+- `telemetry_events_accepted_total`
+- `telemetry_events_rejected_total`
+- `telemetry_events_dropped_total`
+- `telemetry_export_failures_total`
+
+## Missing-data taxonomy
+
+Empty dashboard can mean: no traffic, no failures, broken exporter, wrong query function, wrong environment label, not deployed yet, ingestion delay, datasource/query error.
+
+## False query traps
+
+- Do not query raw URLs; use route templates.
+- Do not group metrics by generated IDs or user hashes.
+- Do not use `rate()` on serverless event gauges; use the documented function for that metric type.

docs/observability/system-overview.md

@@ -0,0 +1,17 @@
+# superset Observability System Overview
+
+Blueprint source: company-wide `observability-blueprint.md`. Vendor stack can vary; principle IDs do not.
+
+Repo: `/Users/aytuncyildizli/superset`
+Surface: agent/desktop/web app surface
+
+## Minimum contract
+
+- Correlation: `request_id`, `trace_id`, `span_id`, `correlation_id`, privacy-safe `actor_id`.
+- Signals: structured logs, metrics, traces/manual business spans, product/client events where applicable.
+- Privacy: no raw secrets, prompts, files, emails, phone numbers, webhook URLs, cookies, or raw customer ids in telemetry.
+- Operations: health and readiness are separate; dashboards/alerts/runbooks live in source control.
+
+## Implementation status
+
+RED until this repo has real helpers/tests/wiring. This directory is the source-control contract seed.

ops/runbooks/observability-triage.md

@@ -0,0 +1,20 @@
+# Runbook: Observability Triage
+
+## First checks
+
+1. Is `/health` live or equivalent process liveness green?
+2. Is `/ready` or dependency readiness green?
+3. Are metrics/logs/traces present for the expected `environment`?
+4. Are recent deploy/version labels visible?
+5. Is missing data caused by no traffic, broken exporter, wrong label, or ingestion delay?
+
+## Privacy guard
+
+Do not paste raw logs if they may contain secrets, prompts, files, emails, phone numbers, webhook URLs, cookies, session strings, or raw customer ids. Redact first.
+
+## Escalation
+
+- SEV1: core journey unavailable, data loss, security/privacy incident.
+- SEV2: major journey degraded for meaningful user segment.
+- SEV3: partial degradation, elevated errors, workaround exists.
+- SEV4: low-impact anomaly/investigation.