@rubytech/create-maxy-code 0.1.314 → 0.1.316

package/payload/platform/lib/graph-search/src/index.ts

@@ -11,15 +11,13 @@
  * The fulltext index `knowledge_fulltext` was renamed to `entity_search`
  * and its label union expanded from 3 labels (KnowledgeDocument | Section |
  * Chunk) to the full operator-meaningful label set (~40 labels) on every
- * textual property the schema's writers assign. Task 416 then split that
- * single `entity_search` into `entity_search_admin` and
- * `entity_search_public` so public BM25 ranking is shaped only by
- * `compiledTruthPublic`, not by the admin-side `compiledTruth` text the
- * public agent can never read. The lib references both via
- * FULLTEXT_INDEX_ADMIN / FULLTEXT_INDEX_PUBLIC below and routes between
- * them with `pickFulltextIndex(allowedScopes)`. The doctrine test at
- * `__tests__/fulltext-coverage.test.ts` parses both declarations and
- * asserts the universal label union plus the compiledTruth field split.
+ * textual property the schema's writers assign. Task 416 split that single
+ * `entity_search` into an admin index plus a public twin; the public twin
+ * and its whole read path were retired once public agents became toolless,
+ * so the lib now targets the single `entity_search_admin` index via
+ * FULLTEXT_INDEX_ADMIN below. The doctrine test at
+ * `__tests__/fulltext-coverage.test.ts` parses that declaration and asserts
+ * the universal label union.
  *
  *   QUERY -->  EMBED -->  VECTOR SEARCH (per index) --> ┐
  *        │                                               ├--> MERGE --> EXPAND --> RESULTS
@@ -83,36 +81,20 @@ function readFlags() {
 export type { DedupLayer, RetrievalClass };
 export { DEFAULT_LAYERS };
 /**
- * Names of the two fulltext indexes (Task 416). Mirrors the `CREATE
- * FULLTEXT INDEX` declarations in `platform/neo4j/schema.cypher`. The
- * doctrine test asserts each name matches its constant and that the field
- * split is correct — drift between schema and these constants breaks
- * BM25 silently.
- *
- * `entity_search_admin` carries `n.compiledTruth`; `entity_search_public`
- * carries `n.compiledTruthPublic`. Other indexed text fields are shared.
+ * Name of the fulltext index (Task 416 admin index; the public twin was
+ * retired with the public read path). Mirrors the `CREATE FULLTEXT INDEX`
+ * declaration in `platform/neo4j/schema.cypher`. The doctrine test asserts
+ * the name matches this constant — drift between schema and this constant
+ * breaks BM25 silently. `entity_search_admin` carries `n.compiledTruth`.
  */
 export const FULLTEXT_INDEX_ADMIN = "entity_search_admin";
-export const FULLTEXT_INDEX_PUBLIC = "entity_search_public";
-
-/**
- * Pick the BM25 index for a read. Returns the public index iff the caller
- * is in the verified public-only scope (`allowedScopes` is exactly
- * `['public']` — the public-agent spawn shape). Anything else — including
- * `undefined` (admin route default), `[]`, `['shared']`, `['public','shared']`
- * — routes to the admin index, so ranking only reflects `compiledTruthPublic`
- * for callers whose Cypher scope filter already restricts them to public
- * rows. Mirrors `isPublicOnlyScope` semantics in memory-search.ts; that
- * file imports this predicate via `isPublicOnlyScope` below.
- */
-export function pickFulltextIndex(
-  allowedScopes: readonly string[] | undefined,
-): string {
-  return isPublicOnlyScope(allowedScopes) ? FULLTEXT_INDEX_PUBLIC : FULLTEXT_INDEX_ADMIN;
-}
 
-/** Detect the public-agent read shape. Hoisted out of memory-search.ts so
- *  the BM25 picker and the property-bag projection share one predicate. */
+/** Detect the public-agent read shape (`allowedScopes` exactly `['public']`).
+ *  The BM25 index picker that used this predicate was retired with the public
+ *  read path. Its one remaining consumer — the `projectPublicProperties` swap
+ *  in the memory MCP's memory-search.ts — is now unreachable: every live read
+ *  passes `allowedScopes: undefined`, so this returns false on every call. Kept
+ *  as that projection's gate for when a public read surface is restored. */
 export function isPublicOnlyScope(
   allowedScopes: readonly string[] | undefined,
 ): boolean {
@@ -481,7 +463,7 @@ export async function bm25Only(
        ORDER BY score DESC
        LIMIT $limit`,
       {
-        indexName: pickFulltextIndex(allowedScopes),
+        indexName: FULLTEXT_INDEX_ADMIN,
         query: escaped,
         accountId,
         limit: int(limit),

package/payload/platform/neo4j/schema.cypher

@@ -429,24 +429,20 @@ OPTIONS {
 //     `role` = discriminator on KnowledgeDocument projections
 //     ('identity'|'soul'|'knowledge'|'knowledge-summary') so BM25 hits surface
 //     which file backed the result. Distinct from Person.role (no shadow).
-// Public/admin BM25 scoring isolation (Task 416). Two fulltext indexes
-// over the same universal label union, differing in one column: the admin
-// index carries `n.compiledTruth`; the public index carries
-// `n.compiledTruthPublic`. The graph-search lib picks between them based
-// on `allowedScopes` — public reads (allowedScopes exactly ['public'])
-// rank against the public twin only; everything else (admin reads pass no
-// scope filter) ranks against the admin field. Before Task 416 a single
-// `entity_search` carried both columns, which let admin notes silently
-// shape the BM25 score of public-twin rows even though the memory-search
-// projection swapped the field on the way out. Drop the pre-416 single
-// index so existing deployments don't keep a populated dead third copy.
+// Admin BM25 fulltext index. Task 416 split a single `entity_search` into an
+// admin index plus a public twin (`entity_search_public`) so a public read
+// ranked only against `n.compiledTruthPublic`. Task 654 retired the public twin
+// and its whole read path once public agents became toolless; `entity_search_admin`
+// is now the only fulltext index, carrying `n.compiledTruth`. Both the pre-416
+// single index and the retired public twin keep a permanent `DROP ... IF EXISTS`
+// below so existing deployments stop populating a dead index that consumes disk.
 DROP INDEX entity_search IF EXISTS;
+DROP INDEX entity_search_public IF EXISTS;
 
-// Task 397 added :ConversationArchive to the label union. DROP + recreate so
-// existing installs pick up the new label without manual reindex; fresh
-// installs no-op the DROP via IF EXISTS.
+// Task 397 added :ConversationArchive to the label union. DROP + recreate the
+// admin index so existing installs pick up new labels without manual reindex;
+// fresh installs no-op the DROP via IF EXISTS.
 DROP INDEX entity_search_admin IF EXISTS;
-DROP INDEX entity_search_public IF EXISTS;
 
 CREATE FULLTEXT INDEX entity_search_admin IF NOT EXISTS
 FOR (n:LocalBusiness|Service|PriceSpecification|OpeningHoursSpecification|Organization
@@ -470,33 +466,11 @@ ON EACH [n.name, n.firstName, n.lastName, n.givenName, n.familyName,
          n.displayName, n.slug, n.role,
          n.client, n.address, n.supplier, n.clientName,
          n.compiledTruth];
-
-CREATE FULLTEXT INDEX entity_search_public IF NOT EXISTS
-FOR (n:LocalBusiness|Service|PriceSpecification|OpeningHoursSpecification|Organization
-    |Listing|Property|Viewing|Offer|PostalAddress
-    |Job|LineItem|Valuation|Milestone|QuoteDocument|VariationNote|InboundInvoice|CheckIn|WhatsAppGroup
-    |Person|UserProfile|Preference|AdminUser|AccessGrant
-    |KnowledgeDocument|ConversationArchive|Section|Chunk|DigitalDocument|CreativeWork|Question|FAQPage|DefinedTerm|Review|ImageObject
-    |Conversation|AdminConversation|PublicConversation|Message|UserMessage|AssistantMessage|ToolCall
-    |Task|Project|Event|Meeting|Attachment
-    |Workflow|WorkflowStep|WorkflowRun|StepResult
-    |EmailAccount
-    |Position|Credential|Agent|Idea|Concept
-    |VoiceProfile|VoiceEdit|SocialPost|AEOAudit|Page
-    |TimelineEvent|CompiledTruthRevision|Report|OrphanCandidate|CitationProposal)
-ON EACH [n.name, n.firstName, n.lastName, n.givenName, n.familyName,
-         n.title, n.currentTitle, n.summary, n.body, n.content, n.text, n.description, n.headline, n.abstract,
-         n.email, n.note, n.label, n.value, n.message, n.preview, n.tagline,
-         n.subject, n.bodyPreview, n.fromName, n.fromAddress, n.agentAddress, n.screeningReason,
-         n.authority, n.contactValue, n.toolName,
-         n.displayName, n.slug, n.role,
-         n.compiledTruthPublic];
 // `compiledTruth` (Task 306) is the authoritative admin-side summary on
-// Person/Company/Concept/Project/LocalBusiness; `compiledTruthPublic`
-// (Task 392) is the customer-facing twin on the four public-twin labels
-// (Organization | Concept | Project | LocalBusiness). Each is indexed in
-// exactly one of the two indexes above, so the BM25 score a query sees
-// is shaped only by the text the caller's scope is allowed to read.
+// Person/Company/Concept/Project/LocalBusiness, indexed above. Its customer-facing
+// twin `compiledTruthPublic` (Task 392) is still written, but is no longer
+// fulltext-indexed: the public index it fed was retired with the public read
+// path (Task 654), so no BM25 surface ranks against it.
 
 // Project node — a standalone creative-output node distinct from
 // :Section. Anchored via (:UserProfile)-[:CREATED]->(:Project), with optional

package/payload/platform/plugins/admin/skills/platform-architecture/SKILL.md

@@ -1,7 +1,7 @@
 ---
 name: platform-architecture
 description: Use when grounding any documented-surface claim about what Maxy ships — plugins, skills, specialists, install/deploy flows, internals. This is the install catalogue, not evidence of what is enabled on the current account. For install state on this account, call `capabilities-here`; for documented surface, cite the `Source:` URL inline.
-content-hash: sha256:6b3de8da9e586c39ea62705d90262664efecf5318cb8b573d7d3912114b5f654
+content-hash: sha256:4c25ba31f010fd3042621f035682a6df66d7ec88011a11984ef197f056a61b60
 brand: maxy-code
 product-name: Maxy
 ---
@@ -2314,7 +2314,7 @@ either is a regression.
 | `/session-rc-spawn` | POST `{ sessionId?, name? }`. Fire-and-forget `claude --remote-control [name] [--session-id <sid>]`. Present `sessionId` resumes; absent starts a fresh session (also used by the sidebar's "New session" button — it no longer opens claude.ai/code directly). Proxies to the manager's `/rc-spawn`, which waits up to **60 s** (raised from 12 s) for the spawned PTY to bind and returns `{ spawnedPid, sessionId, bridgeSessionId, slug, outcome, reason }`. The Sidebar navigates the opened tab to `claude.ai/code/session_<id>` on `outcome=bound`; on `timeout` or `spawn-failed` it shows an error modal (reason + sessionId) and **never** opens a bare claude.ai/code tab. The new process registers itself as its own Remote Control entry in claude.ai/code. | `POST /` |
 | `/claude-sessions` | **Spawn surface only**. `POST /` is the Sidebar new-session-with-prompt path, cookie-auth only (the recorder loopback caller was removed; LinkedIn ingest moved to `/rc-spawn`). The former UI-facing handlers (SSE row feed, list, resume, stop, rename, archive, delete, `/:id/meta`, `/:id/input`, `/:id/log`) were removed — the maxy dashboard no longer manages or displays sessions. | `POST /` |
 
-**Admin session management moved entirely to claude's own interfaces** (claude.ai/code, claude desktop). A manager-owned per-account `claude rc --spawn same-dir` daemon registers the device as a Remote Control target there; the composer creates / resumes / stops / renames / archives / deletes sessions, with model + permission-mode applied at inception. The model lever is `account.json.adminModel` → `CLAUDE_CONFIG_DIR/settings.json "model"`, written by the daemon supervisor at boot; the reasoning-effort lever is `account.json.effort` → `settings.json "effortLevel"` (accepted values `low|medium|high|xhigh`; any other value, e.g. legacy `"auto"`, leaves the key unset so claude's own default governs), written by the same read-merge-write at boot. A third inception lever, permission mode, is added: `account.json.adminPermissionMode` → `settings.json "permissions.defaultMode"` (the 5 composer-writable modes `default|acceptEdits|plan|auto|bypassPermissions`; `dontAsk` is out of scope). Unlike model/effort (top-level keys), this lever **deep-merges** the nested `permissions.defaultMode`, owning only that key and preserving sibling `permissions` keys (allow/deny/…) — what the binary and the spawn lifeline actually read. All three levers are now changeable from the `/chat` composer (the model is a friendly-labelled checkmark dropdown via `MODEL_DISPLAY_NAMES`, effort a Faster↔Smarter slider, mode a 5-mode checkmark dropdown), each writing its `account.json` key then `POST /reapply-levers` so the next spawn honours it; the full write path is in `.docs/admin-webchat-native-channel.md`. Beyond the composer, the maxy admin UI keeps a single "New session" link (`https://claude.ai/code`, opens in a new tab) and no separate session list, viewer, controls, or model/mode picker. The daemon supervisor lives at [`platform/services/claude-session-manager/src/rc-daemon.ts`](../../../services/claude-session-manager/src/rc-daemon.ts). The `/session-defaults` route and `SpawnPreference` node were deleted with the picker. `/new-session-failure`, `/new-session-submit`, and `/claude-capabilities` are now orphaned (consumed only by the deleted NewSessionModal) — see [`.tasks/501`](../../../.tasks/) for their removal.
+**Admin session management moved entirely to claude's own interfaces** (claude.ai/code, claude desktop). A manager-owned per-account `claude rc --spawn same-dir` daemon registers the device as a Remote Control target there; the composer creates / resumes / stops / renames / archives / deletes sessions, with model + permission-mode applied at inception. The model lever is `account.json.adminModel` → `CLAUDE_CONFIG_DIR/settings.json "model"`, written by the daemon supervisor at boot; the reasoning-effort lever is `account.json.effort` → `settings.json "effortLevel"` (accepted values `low|medium|high|xhigh`; any other value, e.g. legacy `"auto"`, leaves the key unset so claude's own default governs), written by the same read-merge-write at boot. A third inception lever, permission mode, is added: `account.json.adminPermissionMode` → `settings.json "permissions.defaultMode"` (the 5 composer-writable modes `default|acceptEdits|plan|auto|bypassPermissions`; `dontAsk` is out of scope). Unlike model/effort (top-level keys), this lever **deep-merges** the nested `permissions.defaultMode`, owning only that key and preserving sibling `permissions` keys (allow/deny/…) — what the binary and the spawn lifeline actually read. All three levers are now changeable from the `/chat` composer (the model is a friendly-labelled checkmark dropdown via `MODEL_DISPLAY_NAMES`, effort a Faster↔Smarter slider, mode a 5-mode checkmark dropdown), each writing its `account.json` key then `POST /reapply-levers` so the next spawn honours it; the full write path is in `.docs/admin-webchat-native-channel.md`. On an existing chat each picker re-seats the live session and carries the operator's current model+mode+effort so picking one never resets the others. When the live session runs in `default` ("Ask permissions") mode, a tool the agent attempts that needs approval surfaces an Allow/Deny card in the composer (tool, description, argument preview); a click sends the verdict over Claude Code's channel permission relay and the tool runs or is refused — webchat is the only interactive ask surface (WhatsApp stays text-only; Telegram has no agent channel). Beyond the composer, the maxy admin UI keeps a single "New session" link (`https://claude.ai/code`, opens in a new tab) and no separate session list, viewer, controls, or model/mode picker. The daemon supervisor lives at [`platform/services/claude-session-manager/src/rc-daemon.ts`](../../../services/claude-session-manager/src/rc-daemon.ts). The `/session-defaults` route and `SpawnPreference` node were deleted with the picker. `/new-session-failure`, `/new-session-submit`, and `/claude-capabilities` are now orphaned (consumed only by the deleted NewSessionModal) — see [`.tasks/501`](../../../.tasks/) for their removal.
 
 **Row title resolution.** Both the sidebar (`/sidebar-sessions`) and the manager's own row payload resolve a row's title in the same order: operator rename → Claude Code `ai-title` → first non-CLI user message → 8-char sessionId prefix. The operator-rename tier is the on-disk `<accountDir>/session-titles.json` (the manager's `UserTitleStore`), keyed by the CC sessionId — the sidebar reads that same file, so a write to the store lights up both surfaces with one write.
 
@@ -2567,8 +2567,8 @@ authoritative. This section names the surfaces and what backs each.
 | `+ New session` button | Opens `NewSessionModal`, which POSTs `{channel, permissionMode, model, initialMessage}` to `/api/admin/claude-sessions`. | `claude-sessions.ts` |
 | Mode trigger | Per-`(accountId, userId)` `SpawnPreference` for `permissionMode` and `model`; persists across reload, tab, device. | `session-defaults.ts` |
 | Nav rows (Chat / People / Agents / Projects / Tasks / Artefacts) | People, Agents, Tasks open the artefact-pane Graph filtered to the matching label. Chat selects the active conversation. Artefacts swaps the list to editable documents. | `graph-search.ts`, `graph-subgraph.ts`, `sidebar-artefacts.ts` |
-| Sessions list (Active / Archived / All) | Live row store driven by SSE; manual reconcile button on the segmented control re-fetches the full id set. Every listed row carries two launch actions: the claude.ai/code resume (`ExternalLink`) and a chat-launch (`MessageSquare`) that opens `/chat?session=<full sessionId>` — the admin webchat pointed at that exact session; sending there resumes a stopped session with the webchat channel bound to that id (see `admin-webchat-native-channel.md`, "Session-targeted mode"). At the ≤720 px drawer breakpoint each row stacks into two lines — dot + title + id/stamp, then the action icons — with a divider between rows and actions at full opacity (the desktop hover-gated 0.5 opacity never resolves on touch). Below a 400 px list width each row's action icons collapse to a single `MoreHorizontal` ellipsis trigger; its popover menu lists the same actions as the inline icons but labels them with concise verbs (`Resume in claude.ai/code`, `Open in webchat`, `Stop`, `Archive` / `Unarchive`, `Rename`, `Delete`) rather than the inline icon buttons' verbose per-session aria-labels — the inline buttons keep those verbose labels because an icon-only control needs the full descriptive accessible name. **Re-seat:** each row also carries a `SessionReseatControl` (sliders icon) beside the action cluster — a model/mode/effort popover that forks the session via `/api/admin/session-reseat` and navigates to the fork (the same fork mechanism `/chat`'s pickers drive). Its model picker defaults to the row's current model (a new `model` field on each row, read from the JSONL tail); mode and effort each carry a "Keep" option that omits the field. | `/claude-sessions/events`, `/claude-sessions`, `/session-reseat` |
-| Channel reader nav rows (WhatsApp / Telegram) | One conditional nav row **per interactive channel that has ≥1 admin session**. Each row carries its brand glyph (WhatsApp `#25D366`, Telegram `#229ED9`); a channel with no admin session shows no row, and no install ever shows an always-empty "No conversations" row. The channel list loads **once on sidebar mount** (the per-channel counts must be known before the nav renders), so a failed load shows one muted, non-clickable line instead of channel rows. Clicking a channel row lists that channel's admin conversations; each conversation row shows a **display name** resolved by precedence `operatorName ?? whatsappName ?? senderId ?? title`: the bound `Person givenName familyName` when a `senderId`→`userId` binding exists, else the persisted WhatsApp display name (`pushName`, read from the latest `:Message:WhatsAppMessage.senderName`), else the raw `senderId`, else the agent title; the agent session title is the hover tooltip only. Under the name each row shows a **last-message-time breadcrumb** (`conv-timestamp`, same markup as the Sessions rows) formatted from the row's `lastMessageAt` (the last parseable turn's `ts`); a session with no parseable turn shows no time, never "Invalid Date". **Re-seat:** each channel conversation row also carries the `SessionReseatControl` beside its click-to-open area (the row is a `conv-with-actions` shell whose open click moved onto an inner button so the popover can sit beside it), so a WhatsApp/Telegram session's model/mode/effort is re-seatable from the dashboard exactly like a webchat row; the channel reader row carries the same `model` field for the picker default. A click on a conversation **opens its transcript on the home surface**: on `/` it selects in place, and from any other route (`/graph`, `/data`, `/browser`) it navigates to `/?wa=<sessionId>&projectDir=<dir>`, which the root shell hydrates on mount (then strips the query so a later refresh does not re-pin a closed conversation). Server logs `[wa-reader] op=operator-name-fallback senderId=… source=whatsapp-pushname` when the pushName fallback fires and `[wa-reader] op=conversations count=<n> withTimestamp=<m>` per list build; the client logs `[admin-ui] wa-open route=… via=<in-place\|navigate> …` and `[admin-ui] wa-hydrate route=/ …`. | `/whatsapp-reader/conversations` |
+| Sessions list (Active / Archived / All) | Live row store driven by SSE; manual reconcile button on the segmented control re-fetches the full id set. Every listed row carries two launch actions: the claude.ai/code resume (`ExternalLink`) and a chat-launch (`MessageSquare`) that opens `/chat?session=<full sessionId>` — the admin webchat pointed at that exact session; sending there resumes a stopped session with the webchat channel bound to that id (see `admin-webchat-native-channel.md`, "Session-targeted mode"). At the ≤720 px drawer breakpoint each row stacks into two lines — dot + title + id/stamp, then the action icons — with a divider between rows and actions at full opacity (the desktop hover-gated 0.5 opacity never resolves on touch). Below a 400 px list width each row's action icons collapse to a single `MoreHorizontal` ellipsis trigger; its popover menu lists the same actions as the inline icons but labels them with concise verbs (`Resume in claude.ai/code`, `Open in webchat`, `Stop`, `Archive` / `Unarchive`, `Rename`, `Delete`, `Re-seat`) rather than the inline icon buttons' verbose per-session aria-labels — the inline buttons keep those verbose labels because an icon-only control needs the full descriptive accessible name. **Re-seat:** Re-seat is a member action of the one `SessionRowActions` cluster, not a separate sibling control — wide it is the sliders icon inline with the other icons (sharing the cluster geometry and alignment); collapsed it is the `Re-seat` menu item, which expands the model/mode/effort form inline inside the overflow menu. The form (`SessionReseatControl`) forks the session via `/api/admin/session-reseat` and navigates to the fork (the same fork mechanism `/chat`'s pickers drive); its model picker defaults to the row's current model (a `model` field on each row, read from the JSONL tail), and mode and effort each carry a "Keep" option that omits the field. The overflow menu and the Re-seat form render through a `document.body` portal, fixed-positioned from the trigger's rect (right-aligned, flipped above the trigger when there is no room below), so neither is clipped by `.side-list`'s `overflow-y:auto`; Escape, a pointerdown outside the trigger/popover, and any scroll or resize dismiss them. | `/claude-sessions/events`, `/claude-sessions`, `/session-reseat` |
+| Channel reader nav rows (WhatsApp / Telegram) | One conditional nav row **per interactive channel that has ≥1 admin session**. Each row carries its brand glyph (WhatsApp `#25D366`, Telegram `#229ED9`); a channel with no admin session shows no row, and no install ever shows an always-empty "No conversations" row. The channel list loads **once on sidebar mount** (the per-channel counts must be known before the nav renders), so a failed load shows one muted, non-clickable line instead of channel rows. Clicking a channel row lists that channel's admin conversations; each conversation row shows a **display name** resolved by precedence `operatorName ?? whatsappName ?? senderId ?? title`: the bound `Person givenName familyName` when a `senderId`→`userId` binding exists, else the persisted WhatsApp display name (`pushName`, read from the latest `:Message:WhatsAppMessage.senderName`), else the raw `senderId`, else the agent title; the agent session title is the hover tooltip only. Under the name each row shows a **last-message-time breadcrumb** (`conv-timestamp`, same markup as the Sessions rows) formatted from the row's `lastMessageAt` (the last parseable turn's `ts`); a session with no parseable turn shows no time, never "Invalid Date". **Re-seat:** each channel conversation row carries a `SessionRowActions` cluster whose sole action is Re-seat (the row is a `conv-with-actions` shell whose open click moved onto an inner button so the cluster can sit beside it), so a WhatsApp/Telegram session's model/mode/effort is re-seatable from the dashboard exactly like a webchat row — wide it is the inline sliders icon, and below the 400 px width it folds into the `…` overflow menu identically to the Sessions rows; the channel reader row carries the same `model` field for the picker default. A click on a conversation **opens its transcript on the home surface**: on `/` it selects in place, and from any other route (`/graph`, `/data`, `/browser`) it navigates to `/?wa=<sessionId>&projectDir=<dir>`, which the root shell hydrates on mount (then strips the query so a later refresh does not re-pin a closed conversation). Server logs `[wa-reader] op=operator-name-fallback senderId=… source=whatsapp-pushname` when the pushName fallback fires and `[wa-reader] op=conversations count=<n> withTimestamp=<m>` per list build; the client logs `[admin-ui] wa-open route=… via=<in-place\|navigate> …` and `[admin-ui] wa-hydrate route=/ …`. | `/whatsapp-reader/conversations` |
 | Channel transcript (`<Transcript>`) | Reads one session's JSONL over SSE (`/whatsapp-reader/stream`) and renders both sides plus tool calls — the turns claude.ai/code hides because channel inbound is stamped `isMeta`. Every turn shows its time-of-day (HH:MM from the turn's `ts`; a null/unparseable `ts` shows no time, never "Invalid Date"). `tool-call`/`tool-result` turns render as **collapsed cards** (chevron + label + time, default collapsed, expandable per card; one card's state never affects another) so the human↔agent text is not buried in JSON; the three conversation kinds always render in full. The thread **follows the tail**: it opens at the newest turn and pins to the bottom on each live append **only while the operator is already within 32 px of the bottom** — a scroll up suppresses the jump so reading history is never yanked. While the operator is scrolled up a **jump-to-bottom control** (`.wa-jump`, bottom-right of the viewport) appears; clicking it scrolls to the newest turn and re-arms follow-tail, and it hides again at the bottom. The thread scrolls inside its grid cell (header/sidebar/footer fixed). The stream sends a 20 s heartbeat so an idle Cloudflare tunnel does not idle-drop, tags each frame with a byte-offset `id:`, and resumes from the client's `Last-Event-ID` on reconnect (no duplicate backlog); the client clears the "Stream disconnected" banner on reopen and latches it only on a terminal CLOSED state. The thread also interleaves a collapsed `⚙ directive injected (N B)` row per turn, sorted by timestamp just before the turn it informed; expanding one fetches the exact stored `prompt-optimiser-directive` bytes for that turn. The entries are listed by `GET /whatsapp-reader/directives?sessionId=` and the bytes served by `GET /whatsapp-reader/directive?sessionId=&name=`, both admin-gated and account-scoped; a missing store degrades to no rows. | `/whatsapp-reader/stream`, `/whatsapp-reader/directives`, `/whatsapp-reader/directive` |
 | Conversations row hover actions | Inline rename, archive, delete, JSONL view / download per row. The historical `.conversations-modal` CSS block exists in `globals.css` but is no longer mounted from any TSX — Sidebar.tsx now owns every per-row affordance directly. | `claude-sessions.ts` |
 | Artefacts list | Lists every `:FileArtifact` under this account's tree (`relativePath STARTS WITH 'accounts/'`, all file types, excluding the `uploads/<id>/` subtree) plus this account's IDENTITY / SOUL / KNOWLEDGE / specialist templates. Click downloads the row's backing file (`downloadPath` → `GET /api/admin/files/download`) so the operator opens it in their local app; rows whose file is outside `DATA_ROOT` (bundled-fallback templates) show a "can't be downloaded" pill. The in-app artefact pane is dead pending removal. | `sidebar-artefacts.ts`, `files.ts` |
@@ -3241,16 +3241,15 @@ The admin agent runs via Claude Code CLI, which manages its own system prompt as
 
 This is assembled as a `<previous-context>` block in the system prompt on each admin turn.
 
-### fetchMemoryContext — the MCP bridge
+### Public-agent memory context (retired)
 
-For public agents, the server calls the memory MCP server via JSON-RPC over stdin/stdout:
-
-1. Spawn the memory MCP server as a subprocess with environment variables: `ACCOUNT_ID`, `ALLOWED_SCOPES=public,shared`, `AGENT_SLUG`, `KNOWLEDGE_KEYWORDS`, `SESSION_ID`
-2. Send `initialize` + `tools/call` (name: `memory-search`, arguments: `{query, account_id}`)
-3. Read the tool result text
-4. Timeout: 8 seconds. On any failure, returns null — the agent proceeds without memory context.
-
-This subprocess model means each public agent query gets an isolated, short-lived memory server instance with the correct scope constraints baked into its environment.
+Public agents historically had their memory context injected by a server-side bridge
+(`fetchMemoryContext`) that spawned the memory MCP server with `ALLOWED_SCOPES=public,shared`
+and ran a `memory-search` on the agent's behalf. That bridge has been removed: public agents
+are now toolless (`buildPublicDeny` denies `mcp__memory__*`), so neither the agent nor a
+server-side proxy runs a public-scope memory read. The public-scope fulltext index that this
+path ranked against was retired alongside it. Restoring read-back for gated visitors is
+tracked separately — see the gated-public-agents doc.
 
 ---
 

package/payload/platform/plugins/docs/references/admin-session.md

@@ -95,7 +95,7 @@ The Hono route `POST /api/admin/claude-sessions` at [`platform/ui/server/routes/
 
 The metadata pane surfaces two distinct identifier values. The three-id model was collapsed back to a single-identity contract: claude's bridge suffix and the JSONL basename UUID are two phases of one claude-side identity, not two operator-visible identifiers. The manager resolver accepts either phase on any route. The wire emits one canonical `sessionId` — whichever phase is current — and the pane shows one row.
 
-**Re-seat mints a new id.** Model, mode, and effort are inception-only levers — claude reads them when a session is born and the `claude rc` daemon exposes no per-session switch — so changing one on an *existing* session forks it: `/api/admin/session-reseat` pre-mints a fresh `sessionId`, the manager `/rc-spawn` fork branch runs `--resume <old> --fork-session --session-id <new> --model <model>` (copying history into the new id), and the operator lands on the fork. The fork pins the chosen model and, when supplied, the chosen `permissionMode` (pushed as `--permission-mode`, one of the 5 composer-writable modes) and `effort` (`low|medium|high|xhigh`, merged into the per-spawn inline `--settings.effortLevel`); each is validated against its allowlist before reaching the argv (a present-but-disallowed value is a 400). Both `/chat`'s composer pickers and the dashboard's per-row Re-seat control drive this one fork, for webchat and WhatsApp/Telegram sessions alike. The full write path is in [`admin-webchat-native-channel.md`](../../../.docs/admin-webchat-native-channel.md).
+**Re-seat mints a new id.** Model, mode, and effort are inception-only levers — claude reads them when a session is born and the `claude rc` daemon exposes no per-session switch — so changing one on an *existing* session forks it: `/api/admin/session-reseat` pre-mints a fresh `sessionId`, the manager `/rc-spawn` fork branch runs `--resume <old> --fork-session --session-id <new> --model <model>` (copying history into the new id), and the operator lands on the fork. The fork pins the chosen model and, when supplied, the chosen `permissionMode` (pushed as `--permission-mode`, one of the 5 composer-writable modes) and `effort` (`low|medium|high|xhigh`, merged into the per-spawn inline `--settings.effortLevel`); each is validated against its allowlist before reaching the argv (a present-but-disallowed value is a 400). Both `/chat`'s composer pickers and the dashboard's per-row Re-seat control drive this one fork, for webchat and WhatsApp/Telegram sessions alike. On `/chat`, every picker now carries the operator's *current* model, mode, and effort and overrides only the picked lever, so changing one never silently resets the other two (a non-writable current mode is omitted rather than rejected). When a re-seated webchat session runs in `default` ("Ask permissions") mode, a tool the agent attempts that needs approval surfaces an Allow/Deny prompt in `/chat` and blocks until the operator answers — over Claude Code's channel permission relay, the only interactive ask surface (WhatsApp stays text-only). The full write path is in [`admin-webchat-native-channel.md`](../../../.docs/admin-webchat-native-channel.md).
 
 | Operator label | What it is | Manager wire field | Log key |
 |---|---|---|---|

package/payload/platform/plugins/docs/references/admin-ui.md

@@ -49,7 +49,7 @@ either is a regression.
 | `/session-rc-spawn` | POST `{ sessionId?, name? }`. Fire-and-forget `claude --remote-control [name] [--session-id <sid>]`. Present `sessionId` resumes; absent starts a fresh session (also used by the sidebar's "New session" button — it no longer opens claude.ai/code directly). Proxies to the manager's `/rc-spawn`, which waits up to **60 s** (raised from 12 s) for the spawned PTY to bind and returns `{ spawnedPid, sessionId, bridgeSessionId, slug, outcome, reason }`. The Sidebar navigates the opened tab to `claude.ai/code/session_<id>` on `outcome=bound`; on `timeout` or `spawn-failed` it shows an error modal (reason + sessionId) and **never** opens a bare claude.ai/code tab. The new process registers itself as its own Remote Control entry in claude.ai/code. | `POST /` |
 | `/claude-sessions` | **Spawn surface only**. `POST /` is the Sidebar new-session-with-prompt path, cookie-auth only (the recorder loopback caller was removed; LinkedIn ingest moved to `/rc-spawn`). The former UI-facing handlers (SSE row feed, list, resume, stop, rename, archive, delete, `/:id/meta`, `/:id/input`, `/:id/log`) were removed — the maxy dashboard no longer manages or displays sessions. | `POST /` |
 
-**Admin session management moved entirely to claude's own interfaces** (claude.ai/code, claude desktop). A manager-owned per-account `claude rc --spawn same-dir` daemon registers the device as a Remote Control target there; the composer creates / resumes / stops / renames / archives / deletes sessions, with model + permission-mode applied at inception. The model lever is `account.json.adminModel` → `CLAUDE_CONFIG_DIR/settings.json "model"`, written by the daemon supervisor at boot; the reasoning-effort lever is `account.json.effort` → `settings.json "effortLevel"` (accepted values `low|medium|high|xhigh`; any other value, e.g. legacy `"auto"`, leaves the key unset so claude's own default governs), written by the same read-merge-write at boot. A third inception lever, permission mode, is added: `account.json.adminPermissionMode` → `settings.json "permissions.defaultMode"` (the 5 composer-writable modes `default|acceptEdits|plan|auto|bypassPermissions`; `dontAsk` is out of scope). Unlike model/effort (top-level keys), this lever **deep-merges** the nested `permissions.defaultMode`, owning only that key and preserving sibling `permissions` keys (allow/deny/…) — what the binary and the spawn lifeline actually read. All three levers are now changeable from the `/chat` composer (the model is a friendly-labelled checkmark dropdown via `MODEL_DISPLAY_NAMES`, effort a Faster↔Smarter slider, mode a 5-mode checkmark dropdown), each writing its `account.json` key then `POST /reapply-levers` so the next spawn honours it; the full write path is in `.docs/admin-webchat-native-channel.md`. Beyond the composer, the maxy admin UI keeps a single "New session" link (`https://claude.ai/code`, opens in a new tab) and no separate session list, viewer, controls, or model/mode picker. The daemon supervisor lives at [`platform/services/claude-session-manager/src/rc-daemon.ts`](../../../services/claude-session-manager/src/rc-daemon.ts). The `/session-defaults` route and `SpawnPreference` node were deleted with the picker. `/new-session-failure`, `/new-session-submit`, and `/claude-capabilities` are now orphaned (consumed only by the deleted NewSessionModal) — see [`.tasks/501`](../../../.tasks/) for their removal.
+**Admin session management moved entirely to claude's own interfaces** (claude.ai/code, claude desktop). A manager-owned per-account `claude rc --spawn same-dir` daemon registers the device as a Remote Control target there; the composer creates / resumes / stops / renames / archives / deletes sessions, with model + permission-mode applied at inception. The model lever is `account.json.adminModel` → `CLAUDE_CONFIG_DIR/settings.json "model"`, written by the daemon supervisor at boot; the reasoning-effort lever is `account.json.effort` → `settings.json "effortLevel"` (accepted values `low|medium|high|xhigh`; any other value, e.g. legacy `"auto"`, leaves the key unset so claude's own default governs), written by the same read-merge-write at boot. A third inception lever, permission mode, is added: `account.json.adminPermissionMode` → `settings.json "permissions.defaultMode"` (the 5 composer-writable modes `default|acceptEdits|plan|auto|bypassPermissions`; `dontAsk` is out of scope). Unlike model/effort (top-level keys), this lever **deep-merges** the nested `permissions.defaultMode`, owning only that key and preserving sibling `permissions` keys (allow/deny/…) — what the binary and the spawn lifeline actually read. All three levers are now changeable from the `/chat` composer (the model is a friendly-labelled checkmark dropdown via `MODEL_DISPLAY_NAMES`, effort a Faster↔Smarter slider, mode a 5-mode checkmark dropdown), each writing its `account.json` key then `POST /reapply-levers` so the next spawn honours it; the full write path is in `.docs/admin-webchat-native-channel.md`. On an existing chat each picker re-seats the live session and carries the operator's current model+mode+effort so picking one never resets the others. When the live session runs in `default` ("Ask permissions") mode, a tool the agent attempts that needs approval surfaces an Allow/Deny card in the composer (tool, description, argument preview); a click sends the verdict over Claude Code's channel permission relay and the tool runs or is refused — webchat is the only interactive ask surface (WhatsApp stays text-only; Telegram has no agent channel). Beyond the composer, the maxy admin UI keeps a single "New session" link (`https://claude.ai/code`, opens in a new tab) and no separate session list, viewer, controls, or model/mode picker. The daemon supervisor lives at [`platform/services/claude-session-manager/src/rc-daemon.ts`](../../../services/claude-session-manager/src/rc-daemon.ts). The `/session-defaults` route and `SpawnPreference` node were deleted with the picker. `/new-session-failure`, `/new-session-submit`, and `/claude-capabilities` are now orphaned (consumed only by the deleted NewSessionModal) — see [`.tasks/501`](../../../.tasks/) for their removal.
 
 **Row title resolution.** Both the sidebar (`/sidebar-sessions`) and the manager's own row payload resolve a row's title in the same order: operator rename → Claude Code `ai-title` → first non-CLI user message → 8-char sessionId prefix. The operator-rename tier is the on-disk `<accountDir>/session-titles.json` (the manager's `UserTitleStore`), keyed by the CC sessionId — the sidebar reads that same file, so a write to the store lights up both surfaces with one write.
 
@@ -302,8 +302,8 @@ authoritative. This section names the surfaces and what backs each.
 | `+ New session` button | Opens `NewSessionModal`, which POSTs `{channel, permissionMode, model, initialMessage}` to `/api/admin/claude-sessions`. | `claude-sessions.ts` |
 | Mode trigger | Per-`(accountId, userId)` `SpawnPreference` for `permissionMode` and `model`; persists across reload, tab, device. | `session-defaults.ts` |
 | Nav rows (Chat / People / Agents / Projects / Tasks / Artefacts) | People, Agents, Tasks open the artefact-pane Graph filtered to the matching label. Chat selects the active conversation. Artefacts swaps the list to editable documents. | `graph-search.ts`, `graph-subgraph.ts`, `sidebar-artefacts.ts` |
-| Sessions list (Active / Archived / All) | Live row store driven by SSE; manual reconcile button on the segmented control re-fetches the full id set. Every listed row carries two launch actions: the claude.ai/code resume (`ExternalLink`) and a chat-launch (`MessageSquare`) that opens `/chat?session=<full sessionId>` — the admin webchat pointed at that exact session; sending there resumes a stopped session with the webchat channel bound to that id (see `admin-webchat-native-channel.md`, "Session-targeted mode"). At the ≤720 px drawer breakpoint each row stacks into two lines — dot + title + id/stamp, then the action icons — with a divider between rows and actions at full opacity (the desktop hover-gated 0.5 opacity never resolves on touch). Below a 400 px list width each row's action icons collapse to a single `MoreHorizontal` ellipsis trigger; its popover menu lists the same actions as the inline icons but labels them with concise verbs (`Resume in claude.ai/code`, `Open in webchat`, `Stop`, `Archive` / `Unarchive`, `Rename`, `Delete`) rather than the inline icon buttons' verbose per-session aria-labels — the inline buttons keep those verbose labels because an icon-only control needs the full descriptive accessible name. **Re-seat:** each row also carries a `SessionReseatControl` (sliders icon) beside the action cluster — a model/mode/effort popover that forks the session via `/api/admin/session-reseat` and navigates to the fork (the same fork mechanism `/chat`'s pickers drive). Its model picker defaults to the row's current model (a new `model` field on each row, read from the JSONL tail); mode and effort each carry a "Keep" option that omits the field. | `/claude-sessions/events`, `/claude-sessions`, `/session-reseat` |
-| Channel reader nav rows (WhatsApp / Telegram) | One conditional nav row **per interactive channel that has ≥1 admin session**. Each row carries its brand glyph (WhatsApp `#25D366`, Telegram `#229ED9`); a channel with no admin session shows no row, and no install ever shows an always-empty "No conversations" row. The channel list loads **once on sidebar mount** (the per-channel counts must be known before the nav renders), so a failed load shows one muted, non-clickable line instead of channel rows. Clicking a channel row lists that channel's admin conversations; each conversation row shows a **display name** resolved by precedence `operatorName ?? whatsappName ?? senderId ?? title`: the bound `Person givenName familyName` when a `senderId`→`userId` binding exists, else the persisted WhatsApp display name (`pushName`, read from the latest `:Message:WhatsAppMessage.senderName`), else the raw `senderId`, else the agent title; the agent session title is the hover tooltip only. Under the name each row shows a **last-message-time breadcrumb** (`conv-timestamp`, same markup as the Sessions rows) formatted from the row's `lastMessageAt` (the last parseable turn's `ts`); a session with no parseable turn shows no time, never "Invalid Date". **Re-seat:** each channel conversation row also carries the `SessionReseatControl` beside its click-to-open area (the row is a `conv-with-actions` shell whose open click moved onto an inner button so the popover can sit beside it), so a WhatsApp/Telegram session's model/mode/effort is re-seatable from the dashboard exactly like a webchat row; the channel reader row carries the same `model` field for the picker default. A click on a conversation **opens its transcript on the home surface**: on `/` it selects in place, and from any other route (`/graph`, `/data`, `/browser`) it navigates to `/?wa=<sessionId>&projectDir=<dir>`, which the root shell hydrates on mount (then strips the query so a later refresh does not re-pin a closed conversation). Server logs `[wa-reader] op=operator-name-fallback senderId=… source=whatsapp-pushname` when the pushName fallback fires and `[wa-reader] op=conversations count=<n> withTimestamp=<m>` per list build; the client logs `[admin-ui] wa-open route=… via=<in-place\|navigate> …` and `[admin-ui] wa-hydrate route=/ …`. | `/whatsapp-reader/conversations` |
+| Sessions list (Active / Archived / All) | Live row store driven by SSE; manual reconcile button on the segmented control re-fetches the full id set. Every listed row carries two launch actions: the claude.ai/code resume (`ExternalLink`) and a chat-launch (`MessageSquare`) that opens `/chat?session=<full sessionId>` — the admin webchat pointed at that exact session; sending there resumes a stopped session with the webchat channel bound to that id (see `admin-webchat-native-channel.md`, "Session-targeted mode"). At the ≤720 px drawer breakpoint each row stacks into two lines — dot + title + id/stamp, then the action icons — with a divider between rows and actions at full opacity (the desktop hover-gated 0.5 opacity never resolves on touch). Below a 400 px list width each row's action icons collapse to a single `MoreHorizontal` ellipsis trigger; its popover menu lists the same actions as the inline icons but labels them with concise verbs (`Resume in claude.ai/code`, `Open in webchat`, `Stop`, `Archive` / `Unarchive`, `Rename`, `Delete`, `Re-seat`) rather than the inline icon buttons' verbose per-session aria-labels — the inline buttons keep those verbose labels because an icon-only control needs the full descriptive accessible name. **Re-seat:** Re-seat is a member action of the one `SessionRowActions` cluster, not a separate sibling control — wide it is the sliders icon inline with the other icons (sharing the cluster geometry and alignment); collapsed it is the `Re-seat` menu item, which expands the model/mode/effort form inline inside the overflow menu. The form (`SessionReseatControl`) forks the session via `/api/admin/session-reseat` and navigates to the fork (the same fork mechanism `/chat`'s pickers drive); its model picker defaults to the row's current model (a `model` field on each row, read from the JSONL tail), and mode and effort each carry a "Keep" option that omits the field. The overflow menu and the Re-seat form render through a `document.body` portal, fixed-positioned from the trigger's rect (right-aligned, flipped above the trigger when there is no room below), so neither is clipped by `.side-list`'s `overflow-y:auto`; Escape, a pointerdown outside the trigger/popover, and any scroll or resize dismiss them. | `/claude-sessions/events`, `/claude-sessions`, `/session-reseat` |
+| Channel reader nav rows (WhatsApp / Telegram) | One conditional nav row **per interactive channel that has ≥1 admin session**. Each row carries its brand glyph (WhatsApp `#25D366`, Telegram `#229ED9`); a channel with no admin session shows no row, and no install ever shows an always-empty "No conversations" row. The channel list loads **once on sidebar mount** (the per-channel counts must be known before the nav renders), so a failed load shows one muted, non-clickable line instead of channel rows. Clicking a channel row lists that channel's admin conversations; each conversation row shows a **display name** resolved by precedence `operatorName ?? whatsappName ?? senderId ?? title`: the bound `Person givenName familyName` when a `senderId`→`userId` binding exists, else the persisted WhatsApp display name (`pushName`, read from the latest `:Message:WhatsAppMessage.senderName`), else the raw `senderId`, else the agent title; the agent session title is the hover tooltip only. Under the name each row shows a **last-message-time breadcrumb** (`conv-timestamp`, same markup as the Sessions rows) formatted from the row's `lastMessageAt` (the last parseable turn's `ts`); a session with no parseable turn shows no time, never "Invalid Date". **Re-seat:** each channel conversation row carries a `SessionRowActions` cluster whose sole action is Re-seat (the row is a `conv-with-actions` shell whose open click moved onto an inner button so the cluster can sit beside it), so a WhatsApp/Telegram session's model/mode/effort is re-seatable from the dashboard exactly like a webchat row — wide it is the inline sliders icon, and below the 400 px width it folds into the `…` overflow menu identically to the Sessions rows; the channel reader row carries the same `model` field for the picker default. A click on a conversation **opens its transcript on the home surface**: on `/` it selects in place, and from any other route (`/graph`, `/data`, `/browser`) it navigates to `/?wa=<sessionId>&projectDir=<dir>`, which the root shell hydrates on mount (then strips the query so a later refresh does not re-pin a closed conversation). Server logs `[wa-reader] op=operator-name-fallback senderId=… source=whatsapp-pushname` when the pushName fallback fires and `[wa-reader] op=conversations count=<n> withTimestamp=<m>` per list build; the client logs `[admin-ui] wa-open route=… via=<in-place\|navigate> …` and `[admin-ui] wa-hydrate route=/ …`. | `/whatsapp-reader/conversations` |
 | Channel transcript (`<Transcript>`) | Reads one session's JSONL over SSE (`/whatsapp-reader/stream`) and renders both sides plus tool calls — the turns claude.ai/code hides because channel inbound is stamped `isMeta`. Every turn shows its time-of-day (HH:MM from the turn's `ts`; a null/unparseable `ts` shows no time, never "Invalid Date"). `tool-call`/`tool-result` turns render as **collapsed cards** (chevron + label + time, default collapsed, expandable per card; one card's state never affects another) so the human↔agent text is not buried in JSON; the three conversation kinds always render in full. The thread **follows the tail**: it opens at the newest turn and pins to the bottom on each live append **only while the operator is already within 32 px of the bottom** — a scroll up suppresses the jump so reading history is never yanked. While the operator is scrolled up a **jump-to-bottom control** (`.wa-jump`, bottom-right of the viewport) appears; clicking it scrolls to the newest turn and re-arms follow-tail, and it hides again at the bottom. The thread scrolls inside its grid cell (header/sidebar/footer fixed). The stream sends a 20 s heartbeat so an idle Cloudflare tunnel does not idle-drop, tags each frame with a byte-offset `id:`, and resumes from the client's `Last-Event-ID` on reconnect (no duplicate backlog); the client clears the "Stream disconnected" banner on reopen and latches it only on a terminal CLOSED state. The thread also interleaves a collapsed `⚙ directive injected (N B)` row per turn, sorted by timestamp just before the turn it informed; expanding one fetches the exact stored `prompt-optimiser-directive` bytes for that turn. The entries are listed by `GET /whatsapp-reader/directives?sessionId=` and the bytes served by `GET /whatsapp-reader/directive?sessionId=&name=`, both admin-gated and account-scoped; a missing store degrades to no rows. | `/whatsapp-reader/stream`, `/whatsapp-reader/directives`, `/whatsapp-reader/directive` |
 | Conversations row hover actions | Inline rename, archive, delete, JSONL view / download per row. The historical `.conversations-modal` CSS block exists in `globals.css` but is no longer mounted from any TSX — Sidebar.tsx now owns every per-row affordance directly. | `claude-sessions.ts` |
 | Artefacts list | Lists every `:FileArtifact` under this account's tree (`relativePath STARTS WITH 'accounts/'`, all file types, excluding the `uploads/<id>/` subtree) plus this account's IDENTITY / SOUL / KNOWLEDGE / specialist templates. Click downloads the row's backing file (`downloadPath` → `GET /api/admin/files/download`) so the operator opens it in their local app; rows whose file is outside `DATA_ROOT` (bundled-fallback templates) show a "can't be downloaded" pill. The in-app artefact pane is dead pending removal. | `sidebar-artefacts.ts`, `files.ts` |

package/payload/platform/plugins/docs/references/internals.md

@@ -373,16 +373,15 @@ The admin agent runs via Claude Code CLI, which manages its own system prompt as
 
 This is assembled as a `<previous-context>` block in the system prompt on each admin turn.
 
-### fetchMemoryContext — the MCP bridge
-
-For public agents, the server calls the memory MCP server via JSON-RPC over stdin/stdout:
-
-1. Spawn the memory MCP server as a subprocess with environment variables: `ACCOUNT_ID`, `ALLOWED_SCOPES=public,shared`, `AGENT_SLUG`, `KNOWLEDGE_KEYWORDS`, `SESSION_ID`
-2. Send `initialize` + `tools/call` (name: `memory-search`, arguments: `{query, account_id}`)
-3. Read the tool result text
-4. Timeout: 8 seconds. On any failure, returns null — the agent proceeds without memory context.
-
-This subprocess model means each public agent query gets an isolated, short-lived memory server instance with the correct scope constraints baked into its environment.
+### Public-agent memory context (retired)
+
+Public agents historically had their memory context injected by a server-side bridge
+(`fetchMemoryContext`) that spawned the memory MCP server with `ALLOWED_SCOPES=public,shared`
+and ran a `memory-search` on the agent's behalf. That bridge has been removed: public agents
+are now toolless (`buildPublicDeny` denies `mcp__memory__*`), so neither the agent nor a
+server-side proxy runs a public-scope memory read. The public-scope fulltext index that this
+path ranked against was retired alongside it. Restoring read-back for gated visitors is
+tracked separately — see the gated-public-agents doc.
 
 ---
 

package/payload/platform/plugins/memory/mcp/dist/index.js

@@ -140,18 +140,15 @@ const validator = {
     liveSource: liveSchemaRuntime.source,
 };
 const userId = process.env.USER_ID; // Optional — present for admin sessions with users.json auth
-// Scope filtering: comma-separated list of allowed scopes (e.g. "public,shared").
-// When set, memory-search only returns nodes whose scope is in this list.
-// When absent (admin), no scope filtering is applied.
-const allowedScopes = process.env.ALLOWED_SCOPES
-    ? process.env.ALLOWED_SCOPES.split(",").map((s) => s.trim()).filter(Boolean)
-    : undefined;
-// Task 485 — per-visitor slice ringfence. Stamped by pty-spawner on
-// gated public-agent spawns. When present, every memory-search WHERE
-// composes `(n.sliceToken = $sliceToken OR n.scope IN $allowedScopes)`
-// so the agent sees its visitor's user-scoped nodes alongside public-scope
-// rows. When absent (admin path + open-mode public), behaviour is
-// unchanged.
+// Scope filtering via `ALLOWED_SCOPES` was retired with the public read path
+// (Task 654): no spawn stamps it any more, so every memory read runs unscoped
+// (admin behaviour). `allowedScopes` is passed as `undefined` below.
+// Task 485 — per-visitor slice ringfence. When `ACCESS_SLICE_TOKEN` is set,
+// every memory-search WHERE composes `(n.sliceToken = $sliceToken OR n.scope
+// IS NULL)` so the agent sees its visitor's user-scoped nodes alongside
+// scope-less rows. The read machinery is retained, but currently inert: no
+// spawn stamps `ACCESS_SLICE_TOKEN` since the public read path was retired
+// (Task 654), so `accessSliceToken` is always undefined and reads run admin-wide.
 const accessSliceToken = typeof process.env.ACCESS_SLICE_TOKEN === "string" && process.env.ACCESS_SLICE_TOKEN.length > 0
     ? process.env.ACCESS_SLICE_TOKEN
     : undefined;
@@ -259,7 +256,7 @@ eagerTool(server, "memory-search", "Search the knowledge graph using semantic ve
             accountId,
             limit,
             expandHops,
-            allowedScopes,
+            allowedScopes: undefined,
             sliceToken: accessSliceToken,
             keywords,
             keywordMatch,
@@ -368,7 +365,7 @@ eagerTool(server, "memory-lookup-by-name", "Deterministic by-name entity lookup.
             name,
             accountId,
             labels,
-            allowedScopes,
+            allowedScopes: undefined,
             sliceToken: accessSliceToken,
             agentSlug: effectiveAgentSlug,
             limit,
@@ -376,7 +373,7 @@ eagerTool(server, "memory-lookup-by-name", "Deterministic by-name entity lookup.
         process.stderr.write(`[memory-lookup-by-name] name=${JSON.stringify(name)} ` +
             `labels=${labels && labels.length > 0 ? labels.join("|") : "any"} ` +
             `resultCount=${hits.length} ` +
-            `scope=${accessSliceToken ? accessSliceToken.slice(0, 8) : allowedScopes ? allowedScopes.join("|") : "admin"}\n`);
+            `scope=${accessSliceToken ? accessSliceToken.slice(0, 8) : "admin"}\n`);
         if (hits.length === 0) {
             return {
                 content: [