@agent-native/core 0.98.2 → 0.98.4

package/corpus/core/src/demo/redact.ts

@@ -1,17 +1,26 @@
 /**
  * Pure, dependency-free, deterministic demo-mode redactor.
  *
- * Replaces sensitive values (names, emails, and free numbers) with stable fake
+ * Replaces sensitive values (emails and free numbers) with stable fake
  * substitutes so a demo looks coherent — the same input always maps to the
  * same fake. Crucially, it NEVER rewrites identifiers, structural tokens, or
- * timestamps: under-redaction is safe, corrupting an ID is not. The string
- * redactor uses a protect-first strategy (mask IDs with opaque placeholders
- * before any transform runs, restore them byte-identical afterwards), and the
- * structure-aware walker additionally protects leaf values by key name.
+ * timestamps. Names and other free text are deliberately left alone because
+ * guessing whether arbitrary text is a person's name is too inaccurate. The
+ * string redactor uses a protect-first strategy (mask IDs with opaque
+ * placeholders before any transform runs, restore them byte-identical
+ * afterwards), and the structure-aware walker additionally protects leaf
+ * values by key name.
  */
 
 export interface RedactOptions {
   salt?: string;
+  /** Redact numeric values. Defaults to true for backward compatibility. */
+  redactNumbers?: boolean;
+  /**
+   * Redact emails stored in otherwise protected structural fields such as
+   * `userId` and `userKey`. Intended for display-only frontend responses.
+   */
+  redactProtectedEmails?: boolean;
 }
 
 /* ------------------------------------------------------------------ *
@@ -66,7 +75,7 @@ function pick<T>(rng: () => number, pool: readonly T[]): T {
  *      round-trips back through redaction (e.g. you edit a draft that's
  *      already showing fake text, it autosaves, then refetches) it is passed
  *      through UNCHANGED instead of being re-faked into something new. This
- *      is what stops names/emails drifting on every edit.
+ *      is what stops emails drifting on every edit.
  *
  * Leak-free by construction: no timers, a hard size cap, and lazy TTL purge
  * on write. Works the same per-tab in the browser and per-process on the
@@ -242,15 +251,6 @@ const LAST_NAMES: readonly string[] = [
  * Fake value generators (deterministic in value + salt)
  * ------------------------------------------------------------------ */
 
-function fakeName(original: string, salt: string): string {
-  return memoFake("name", original, salt, () => {
-    const rng = seededRng(`name:${original}`, salt);
-    const first = pick(rng, FIRST_NAMES);
-    const last = pick(rng, LAST_NAMES);
-    return `${first} ${last}`;
-  });
-}
-
 // Realistic-looking but safe stand-in domains: real consumer providers (the
 // local part is fake, so no real mailbox is implied) plus the canonical
 // reserved fictional-company domains (Contoso/Fabrikam/Northwind), which read
@@ -484,10 +484,6 @@ function unprotect(text: string, restore: Map<string, string>): string {
 
 const EMAIL_RE = /\b[A-Za-z0-9._%+-]+@[A-Za-z0-9.-]+\.[A-Za-z]{2,}\b/g;
 
-// 2+ capitalized words (each starts uppercase, allows internal lowercase /
-// apostrophes / hyphens). Single-letter middle initials allowed: `Sarah J Connor`.
-const FULL_NAME_RE = /\b[A-Z][a-zA-Z'’-]*(?:\s+[A-Z](?:[a-zA-Z'’-]*)?){1,3}\b/g;
-
 // A standalone numeric token: optional currency + sign, digits with optional
 // comma grouping and a single optional decimal part. Bounded so it is NOT
 // adjacent to a letter (placeholders already removed letter-mixed tokens, but
@@ -499,10 +495,6 @@ function transformEmails(text: string, salt: string): string {
   return text.replace(EMAIL_RE, (match) => fakeEmail(match, salt));
 }
 
-function transformNames(text: string, salt: string): string {
-  return text.replace(FULL_NAME_RE, (match) => fakeName(match, salt));
-}
-
 function isProbablyYear(digits: string): boolean {
   if (digits.length !== 4) return false;
   const n = Number(digits);
@@ -550,20 +542,26 @@ function transformNumbers(text: string, salt: string): string {
 function redactDemoStringInternal(
   text: string,
   salt: string,
-  redactNames: boolean,
+  redactNumbers: boolean,
 ): string {
   if (typeof text !== "string" || text.length === 0) return text;
+  if (!text.includes("@") && (!redactNumbers || !/[\d$€£¥]/.test(text))) {
+    return text;
+  }
 
   const { text: masked, restore } = protect(text);
   let out = masked;
   out = transformEmails(out, salt);
-  if (redactNames) out = transformNames(out, salt);
-  out = transformNumbers(out, salt);
+  if (redactNumbers) out = transformNumbers(out, salt);
   return unprotect(out, restore);
 }
 
 export function redactDemoString(text: string, opts?: RedactOptions): string {
-  return redactDemoStringInternal(text, opts?.salt ?? "", true);
+  return redactDemoStringInternal(
+    text,
+    opts?.salt ?? "",
+    opts?.redactNumbers !== false,
+  );
 }
 
 /* ------------------------------------------------------------------ *
@@ -573,35 +571,14 @@ export function redactDemoString(text: string, opts?: RedactOptions): string {
 // Keys whose leaf values must NEVER be transformed (still recurse into nested
 // objects/arrays under them). Besides ids/urls/timestamps this also covers
 // code/query-bearing keys (`sql`, `query`, `expression`, `formula`, `code`):
-// prose-redacting a SQL string injects fake emails/numbers into it and rewrites
-// its leading tokens, producing `syntax error at or near "Henry"` /
-// `queries must start with SELECT or WITH`. The query must run untouched; its
-// RESULTS are what get redacted so the chart still shows fake values.
+// redacting a SQL string mutates literals and can make the query semantically
+// wrong or invalid. The query must run untouched; its RESULTS are what get
+// redacted so the chart still shows fake values.
 const PROTECTED_KEY_RE =
   /^id$|(^|_)id$|Id$|Ids$|uuid|guid|slug|token|secret|password|passwd|apikey|api_key|hash|sha\d*|etag|cursor|nonce|sessionid|messageid|threadid|nodeid|(^|_)key$|keyid|(^|_)ref$|url$|uri$|href$|src$|path$|filename$|mimetype|mime|^sql$|sql$|query|expression|formula|^code$|createdat|updatedat|deletedat|expiresat|timestamp|.+at$|.+_at$/i;
 
-// UI labels and chart/dashboard titles often look like person names ("Maya
-// Davis (First-party)", "Clicks by Henry Moore") but are structural labels. Keep
-// them stable while still redacting free-text and contact-style person fields.
-const LABEL_TEXT_KEY_RE = /^(name|title|label)$/i;
-
 const MAX_DEPTH = 64;
 
-function isStandalonePersonNameLabel(text: string): boolean {
-  const trimmed = text.trim();
-  if (!trimmed) return false;
-  return /^[A-Z][a-zA-Z'’-]+(?:\s+[A-Z]\.?)?\s+[A-Z][a-zA-Z'’-]+$/.test(
-    trimmed,
-  );
-}
-
-function shouldRedactNamesForChild(key: string, entry: unknown): boolean {
-  if (!LABEL_TEXT_KEY_RE.test(key)) return true;
-  // Preserve structural labels/titles, but do not let common contact-shaped
-  // fields like `{ name: "Olivia Parker" }` leak real person names.
-  return typeof entry === "string" ? isStandalonePersonNameLabel(entry) : true;
-}
-
 function isPlainObject(value: unknown): value is Record<string, unknown> {
   if (value === null || typeof value !== "object") return false;
   if (Array.isArray(value)) return false;
@@ -610,14 +587,6 @@ function isPlainObject(value: unknown): value is Record<string, unknown> {
   return proto === Object.prototype || proto === null;
 }
 
-function redactDemoStringWithSalt(
-  text: string,
-  salt: string,
-  redactNames: boolean,
-): string {
-  return redactDemoStringInternal(text, salt, redactNames);
-}
-
 function redactNumberLeaf(value: number, salt: string): number {
   if (!Number.isFinite(value)) return value;
   const repr = String(value);
@@ -635,7 +604,8 @@ function walk(
   salt: string,
   depth: number,
   seen: WeakSet<object>,
-  redactNames: boolean,
+  redactNumbers: boolean,
+  redactProtectedEmails: boolean,
 ): unknown {
   if (depth > MAX_DEPTH) return value;
 
@@ -644,11 +614,11 @@ function walk(
   const t = typeof value;
 
   if (t === "string") {
-    return redactDemoStringWithSalt(value as string, salt, redactNames);
+    return redactDemoStringInternal(value as string, salt, redactNumbers);
   }
 
   if (t === "number") {
-    return redactNumberLeaf(value as number, salt);
+    return redactNumbers ? redactNumberLeaf(value as number, salt) : value;
   }
 
   if (t === "boolean" || t === "bigint" || t === "function" || t === "symbol") {
@@ -661,7 +631,7 @@ function walk(
     if (seen.has(value)) return value;
     seen.add(value);
     const out = value.map((item) =>
-      walk(item, salt, depth + 1, seen, redactNames),
+      walk(item, salt, depth + 1, seen, redactNumbers, redactProtectedEmails),
     );
     seen.delete(value);
     return out;
@@ -681,7 +651,20 @@ function walk(
         ) {
           // Still recurse into nested structures, but the protected key does
           // not transform its own leaf value.
-          out[key] = walk(entry, salt, depth + 1, seen, redactNames);
+          out[key] = walk(
+            entry,
+            salt,
+            depth + 1,
+            seen,
+            redactNumbers,
+            redactProtectedEmails,
+          );
+        } else if (
+          redactProtectedEmails &&
+          typeof entry === "string" &&
+          entry.includes("@")
+        ) {
+          out[key] = redactDemoStringInternal(entry, salt, false);
         } else {
           // Leaf under a protected key: pass through completely untouched.
           out[key] = entry;
@@ -693,7 +676,8 @@ function walk(
         salt,
         depth + 1,
         seen,
-        redactNames && shouldRedactNamesForChild(key, entry),
+        redactNumbers,
+        redactProtectedEmails,
       );
     }
     seen.delete(value);
@@ -706,7 +690,14 @@ function walk(
 
 export function redactDemoData<T>(value: T, opts?: RedactOptions): T {
   const salt = opts?.salt ?? "";
-  return walk(value, salt, 0, new WeakSet<object>(), true) as T;
+  return walk(
+    value,
+    salt,
+    0,
+    new WeakSet<object>(),
+    opts?.redactNumbers !== false,
+    opts?.redactProtectedEmails === true,
+  ) as T;
 }
 
 /**

package/corpus/templates/analytics/.agents/skills/session-replay/SKILL.md

@@ -102,12 +102,60 @@ agent answers about browser recordings in the Analytics template.
 - Wait for all replay chunks (`isComplete`) before constructing the rrweb
   `Replayer`. Progressive chunk publishes should only update the loading bar;
   rebuilding the player mid-load desyncs the scrubber and playhead.
-- Pass events to `Replayer` untouched. rrweb rebuilds them in a sandboxed iframe;
-  pre-processing DOM, stylesheet, resource, or mutation payloads makes playback
-  diverge from the captured page. Let rrweb own iframe sizing via Meta /
-  ViewportResize, and keep the outer wrapper in that same coordinate system for
-  fit-to-stage scaling. Never clamp only one layer, rewrite Meta widths, or force
-  iframe dimensions — that desyncs the FullSnapshot DOM and breaks the stage.
+- Pass normal events to `Replayer` untouched. rrweb rebuilds them in a sandboxed
+  iframe; pre-processing DOM, stylesheet, resource, or mutation payloads makes
+  playback diverge from the captured page. In particular, never rewrite `href`, `src`,
+  `_cssText`, CSS `url()`, or Meta URLs to `about:blank`; that exact remediation
+  broke historical replay CSS in PR #2040. Handle request privacy at capture or
+  the sandbox boundary instead of mutating stored rrweb events. Historical
+  captures without inlined resources require live stylesheet/image/font
+  requests for accurate rendering; the viewer accepts that fidelity tradeoff,
+  uses rrweb's script-disabled sandbox plus `referrerpolicy="no-referrer"`, and
+  must never add credentials or proxy those URLs through a privileged server.
+- Capture-time URL scrubbing must preserve load-bearing DOM resource attributes:
+  `src`, `srcset`, `poster`, `data`, and `href` only on resource links such as
+  stylesheets, preloads, and icons. Signed CDN query parameters are part of the
+  resource identity; redacting them produces missing CSS, fonts, images, and
+  oversized fallback icons. Keep scrubbing Meta/navigation URLs, anchor hrefs,
+  and console/network diagnostics. Captured `_cssText` and CSS `@import`/`url()`
+  values must remain byte-identical.
+- rrweb rebuilds into an `about:srcdoc` iframe, which inherits the Analytics
+  document's CSP. Analytics currently sends no CSP header; if a future change
+  adds restrictive `style-src`, `font-src`, or `img-src` directives, verify
+  historical replays and resolve external imports/fonts at capture before
+  blocking the recorded resource origins. Do not diagnose current font loss as
+  CSP without checking the deployed response headers first.
+- Let rrweb own normal iframe sizing via Meta / ViewportResize, and keep the
+  outer wrapper in that same coordinate system for fit-to-stage scaling. Some
+  legacy recordings contain demonstrably corrupt 4,000–7,500px-wide viewport
+  values from ordinary desktop sessions. Only widths of at least 4,000px with
+  aspect ratios above 4:1 are recovered to a 16:10 viewport; real 32:9 and
+  mobile portrait captures remain untouched. The resize handler applies the
+  recovery to **both** rrweb's iframe and the outer stage. This includes the
+  historical 3,000–3,999px-wide, sub-1,000px-high malformed shape; do not
+  narrow the guard back to only 4,000px+ recordings.
+  Never remove this recovery, clamp only one layer, or broadly rewrite event
+  geometry: those changes repeatedly recreated the ribbon/clipping regression.
+- The malformed-viewport recovery's sole event-data exception is pointer
+  geometry: project MouseMove, TouchMove, Drag, and MouseInteraction x/y into
+  the corrected camera and clamp impossible coordinates. Correcting the iframe
+  without its paired pointer projection makes rrweb cast valid target ids
+  thousands of pixels off-stage. Normal-viewport pointer events must retain
+  their original object references and exact coordinates.
+- Keep rrweb's stock cursor stylesheet and its hotspot transform. During
+  playback, hide the viewer's native pointer over Analytics' transparent
+  click-to-pause overlay so it cannot masquerade as a frozen recorded cursor.
+- Keep rrweb's recorded focus handling enabled. Focus and focus-visible state
+  affect menus, forms, and keyboard UX; disabling `triggerFocus` makes a valid
+  snapshot diverge from the source page.
+- `insertStyleRules` may suppress known toast/snackbar containers only. Never
+  hide generic framework primitives such as
+  `[data-radix-popper-content-wrapper]`: Radix dropdowns, selects, tooltips,
+  and other real recorded product UI all share that wrapper.
+- Keep the realistic fidelity and malformed-viewport tests in
+  `SessionDetailPage.spec.ts`. Do not change their expectations merely to bless
+  a new sanitizer or raw ultra-wide sizing; validate the affected replay in a
+  browser first.
 - The event timeline soft-highlights the active marker, auto-scrolls it into
   view (pausing briefly after manual scroll), and supports search. It appears
   beside the player from ~880px content width upward.
@@ -144,3 +192,14 @@ agent answers about browser recordings in the Analytics template.
   `.an-mask` or `data-an-mask`.
 - Use `.an-block`, `.an-ignore`, `data-an-block`, or `data-an-ignore` for
   sensitive zones that should not be captured.
+- A definitive upload `409` abandons only the conflicted replay identity and
+  immediately starts rrweb again under a fresh per-tab id, producing a new
+  Meta + FullSnapshot for long-lived SPA tabs. Recovery is limited to one
+  restart until an upload succeeds so a misconfigured endpoint cannot loop;
+  Analytics tracks the content-free `session replay upload rejected` lifecycle
+  event so conflicts and recovery success are measurable.
+- Do not label an old recording "corrupt" from pointer coordinates, unknown
+  mutation node ids, or changing Meta geometry alone. Those shapes can be
+  legitimate with scrolling, iframes/shadow DOM, navigation, and resize. A
+  historical-artifact notice needs a durable capture/ingest marker or another
+  low-false-positive invariant; do not guess from playback heuristics.

package/corpus/templates/analytics/AGENTS.md

@@ -41,12 +41,12 @@ details live in `.agents/skills/`.
   unless the user explicitly asks to inspect the demo dashboard.
 - Framework Demo mode is separate from the built-in `demo` source. While Demo
   mode is enabled, Analytics line and area charts reshape each returned numeric
-  series into a deterministic, natural-looking upward trend using that series'
-  actual minimum, maximum, and normalized volatility. Smooth source series stay
-  smooth; spiky source series keep larger swings. This is presentation-only:
-  query results, tables, metrics, exports, and non-time-series charts remain
-  unchanged, so do not cite the displayed intermediate line/area points as
-  retrieved source values.
+  series into a deterministic upward trend using that series' actual minimum,
+  maximum, and local step pattern. Source spikes and dips remain at the same
+  x-positions while the smallest necessary linear drift moves the series up and
+  to the right. This is presentation-only: query results, tables, metrics,
+  exports, and non-time-series charts remain unchanged, so do not cite the
+  displayed intermediate line/area points as retrieved source values.
 - Every analytical answer should include enough audit context for the user to
   trust it: source(s), time window, filters, sample size or row count,
   join/match method when relevant, and caveats/gaps.