refactor(chunks): append-only chunk log with per-step cache-stable wire

Replace the message-as-container model with a flat, append-only chunk log.

- chunks table (id, tab_id, seq, turn_id, step, role, type, data_json): one
  row per chunk; tool_call (assistant) and tool_result (tool) are SEPARATE
  rows linked by callId. Message/turn are derived groupings, not stored.
- chunks/transform.ts: DB-free explode (Chunk[] -> rows) / group (rows ->
  messages), shared by backend and the browser frontend.
- Cache fix: toModelMessages segments each turn at tool-batch boundaries into
  stable [assistant, tool] pairs per step, so earlier steps serialize
  byte-identically across requests (kills the prompt-cache churn).
- agent-manager persists a turn's chunks on seal (once), discarding a failed
  fallback attempt's partial chunks; rebuilds agent history from the log.
- GET /messages windows the log by chunk seq then groups; loadMoreMessages
  merges a turn split across the window boundary by turnId.
- One-shot migration drops the legacy messages table and clears tabs;
  settings/credentials/keys/usage preserved.

Full suite green (317 tests); biome, tsc, and svelte-check clean.

7 files changed, 543 insertions, 169 deletions

diff --git a/packages/core/src/agent/agent.ts b/packages/core/src/agent/agent.ts
index 439b436..4638301 100644
--- a/packages/core/src/agent/agent.ts
+++ b/packages/core/src/agent/agent.ts
diff --git a/packages/core/src/chunks/transform.ts b/packages/core/src/chunks/transform.ts
new file mode 100644
index 0000000..a4c6fc8
--- /dev/null
+++ b/packages/core/src/chunks/transform.ts
@@ -0,0 +1,266 @@
+// Pure, dependency-free transforms between the render-shaped `Chunk[]` /
+// `ChatMessage` model and the flat append-only `ChunkRow` log. Kept free of any
+// DB (`bun:sqlite`) import so BOTH the backend persistence layer
+// (`db/chunks.ts`) and the browser frontend store can share the exact same
+// explode/group logic.
+
+import type {
+	Chunk,
+	ChunkRow,
+	ChunkRowDraft,
+	ErrorData,
+	MessageRole,
+	SystemData,
+	TextData,
+	ThinkingData,
+	ToolBatchChunk,
+	ToolCallData,
+	ToolResultData,
+} from "../types/index.js";
+
+/**
+ * A DERIVED message — a grouping of contiguous chunk rows reconstructed for the
+ * agent's in-memory history and for the frontend's render bubbles. NOT a stored
+ * shape: the source of truth is the flat chunk log.
+ */
+export interface MessageRow {
+	id: string;
+	tabId: string;
+	seq: number;
+	/** turn_id of the chunk rows this message was grouped from. */
+	turnId: string;
+	role: MessageRole;
+	chunks: Chunk[];
+	createdAt: number;
+}
+
+// ─── Explode: in-memory turn → flat chunk-row drafts ─────────────
+//
+// A turn's render-shaped `Chunk[]` is flattened into append-only rows.
+// `tool-batch` chunks are split into SEPARATE `tool_call` (role=assistant) and
+// `tool_result` (role=tool) rows linked by `callId`, mapping 1:1 to the
+// Anthropic wire format. `step` increments after each tool-batch: every LLM
+// round-trip emits text/thinking then (optionally) one tool-batch, so a
+// tool-batch boundary is exactly a step boundary.
+
+/** Explode a single user message's text into one row draft. */
+export function explodeUserText(turnId: string, text: string): ChunkRowDraft[] {
+	return [{ turnId, step: 0, role: "user", type: "text", data: { text } }];
+}
+
+/** Explode an assistant turn's accumulated chunks into ordered row drafts. */
+export function explodeTurn(turnId: string, chunks: Chunk[]): ChunkRowDraft[] {
+	const drafts: ChunkRowDraft[] = [];
+	let step = 0;
+	for (const chunk of chunks) {
+		switch (chunk.type) {
+			case "text":
+				drafts.push({ turnId, step, role: "assistant", type: "text", data: { text: chunk.text } });
+				break;
+			case "thinking":
+				drafts.push({
+					turnId,
+					step,
+					role: "assistant",
+					type: "thinking",
+					data: {
+						text: chunk.text,
+						...(chunk.metadata !== undefined ? { metadata: chunk.metadata } : {}),
+					},
+				});
+				break;
+			case "tool-batch": {
+				for (const call of chunk.calls) {
+					drafts.push({
+						turnId,
+						step,
+						role: "assistant",
+						type: "tool_call",
+						data: { callId: call.id, name: call.name, arguments: call.arguments },
+					});
+				}
+				for (const call of chunk.calls) {
+					if (call.result === undefined) continue;
+					drafts.push({
+						turnId,
+						step,
+						role: "tool",
+						type: "tool_result",
+						data: {
+							callId: call.id,
+							name: call.name,
+							result: call.result,
+							isError: call.isError ?? false,
+							...(call.shellOutput !== undefined ? { shellOutput: call.shellOutput } : {}),
+						},
+					});
+				}
+				// A tool-batch ends the current LLM step; subsequent chunks belong
+				// to the next round-trip.
+				step++;
+				break;
+			}
+			case "error":
+				drafts.push({
+					turnId,
+					step,
+					role: "assistant",
+					type: "error",
+					data: {
+						message: chunk.message,
+						...(chunk.statusCode !== undefined ? { statusCode: chunk.statusCode } : {}),
+					},
+				});
+				break;
+			case "system":
+				drafts.push({
+					turnId,
+					step,
+					role: "system",
+					type: "system",
+					data: { kind: chunk.kind, text: chunk.text },
+				});
+				break;
+		}
+	}
+	return drafts;
+}
+
+// ─── Group: flat chunk rows → derived render messages ────────────
+//
+// The inverse of explode (best-effort over an arbitrary window, so it tolerates
+// orphan tool-results whose tool-call was paged out). `tool_result` rows
+// (role=tool) merge back into the preceding assistant message's per-step
+// `tool-batch` chunk by `callId` rather than forming their own message.
+
+export function groupRowsToMessages(rows: ChunkRow[]): MessageRow[] {
+	const messages: MessageRow[] = [];
+
+	let current: { msg: MessageRow; batches: Map<number, ToolBatchChunk> } | null = null;
+	const flush = () => {
+		if (current) {
+			messages.push(current.msg);
+			current = null;
+		}
+	};
+	const ensureAssistant = (row: ChunkRow) => {
+		if (!current) {
+			current = {
+				msg: {
+					id: row.id,
+					tabId: row.tabId,
+					seq: row.seq,
+					turnId: row.turnId,
+					role: "assistant",
+					chunks: [],
+					createdAt: row.createdAt,
+				},
+				batches: new Map(),
+			};
+		}
+		return current;
+	};
+	const ensureBatch = (step: number): ToolBatchChunk => {
+		const c = current;
+		if (!c) throw new Error("ensureBatch called without an assistant message");
+		let batch = c.batches.get(step);
+		if (!batch) {
+			batch = { type: "tool-batch", calls: [] };
+			c.batches.set(step, batch);
+			c.msg.chunks.push(batch);
+		}
+		return batch;
+	};
+
+	for (const row of rows) {
+		if (row.role === "user") {
+			flush();
+			const d = row.data as TextData;
+			messages.push({
+				id: row.id,
+				tabId: row.tabId,
+				seq: row.seq,
+				turnId: row.turnId,
+				role: "user",
+				chunks: [{ type: "text", text: d.text }],
+				createdAt: row.createdAt,
+			});
+			continue;
+		}
+		if (row.role === "system") {
+			// Coalesce consecutive system rows into one system message (multiple
+			// system chunks), matching the old applySystemEvent behaviour.
+			const prev = messages[messages.length - 1];
+			const d = row.data as SystemData;
+			if (current === null && prev && prev.role === "system") {
+				prev.chunks.push({ type: "system", kind: d.kind, text: d.text });
+				continue;
+			}
+			flush();
+			messages.push({
+				id: row.id,
+				tabId: row.tabId,
+				seq: row.seq,
+				turnId: row.turnId,
+				role: "system",
+				chunks: [{ type: "system", kind: d.kind, text: d.text }],
+				createdAt: row.createdAt,
+			});
+			continue;
+		}
+
+		// assistant / tool rows → part of the current assistant message
+		const c = ensureAssistant(row);
+		switch (row.type) {
+			case "text":
+				c.msg.chunks.push({ type: "text", text: (row.data as TextData).text });
+				break;
+			case "thinking": {
+				const d = row.data as ThinkingData;
+				c.msg.chunks.push({
+					type: "thinking",
+					text: d.text,
+					...(d.metadata !== undefined ? { metadata: d.metadata } : {}),
+				});
+				break;
+			}
+			case "error": {
+				const d = row.data as ErrorData;
+				c.msg.chunks.push({
+					type: "error",
+					message: d.message,
+					...(d.statusCode !== undefined ? { statusCode: d.statusCode } : {}),
+				});
+				break;
+			}
+			case "tool_call": {
+				const d = row.data as ToolCallData;
+				ensureBatch(row.step).calls.push({ id: d.callId, name: d.name, arguments: d.arguments });
+				break;
+			}
+			case "tool_result": {
+				const d = row.data as ToolResultData;
+				const batch = ensureBatch(row.step);
+				const entry = batch.calls.find((e) => e.id === d.callId);
+				if (entry) {
+					entry.result = d.result;
+					entry.isError = d.isError;
+					if (d.shellOutput !== undefined) entry.shellOutput = d.shellOutput;
+				} else {
+					// Orphan result (its tool_call was paged out of this window).
+					batch.calls.push({
+						id: d.callId,
+						name: d.name,
+						arguments: {},
+						result: d.result,
+						isError: d.isError,
+						...(d.shellOutput !== undefined ? { shellOutput: d.shellOutput } : {}),
+					});
+				}
+				break;
+			}
+		}
+	}
+	flush();
+	return messages;
+}
diff --git a/packages/core/src/db/chunks.ts b/packages/core/src/db/chunks.ts
new file mode 100644
index 0000000..6841eb5
--- /dev/null
+++ b/packages/core/src/db/chunks.ts
@@ -0,0 +1,150 @@
+import { randomUUID } from "node:crypto";
+import {
+	explodeTurn,
+	explodeUserText,
+	groupRowsToMessages,
+	type MessageRow,
+} from "../chunks/transform.js";
+import type { ChunkData, ChunkRow, ChunkRowDraft, TextData } from "../types/index.js";
+import { getDatabase } from "./index.js";
+
+// Re-export the DB-free transforms so existing barrel consumers
+// (`@dispatch/core`) keep importing them from here. The browser frontend deep-
+// imports them directly from `chunks/transform.js` to avoid the DB dependency.
+export { explodeTurn, explodeUserText, groupRowsToMessages, type MessageRow };
+
+// ─── Persistence ─────────────────────────────────────────────────
+
+function mapRow(row: Record<string, unknown>): ChunkRow {
+	let data: ChunkData;
+	try {
+		data = JSON.parse(row.data_json as string) as ChunkData;
+	} catch {
+		data = { text: "" } as TextData;
+	}
+	return {
+		id: row.id as string,
+		tabId: row.tab_id as string,
+		seq: row.seq as number,
+		turnId: row.turn_id as string,
+		step: row.step as number,
+		role: row.role as ChunkRow["role"],
+		type: row.type as ChunkRow["type"],
+		data,
+		createdAt: row.created_at as number,
+	};
+}
+
+/**
+ * Append one or more chunk-row drafts to a tab, assigning a monotonic per-tab
+ * `seq` and a fresh id/timestamp to each. Returns the inserted rows in order.
+ */
+export function appendChunks(tabId: string, drafts: ChunkRowDraft[]): ChunkRow[] {
+	if (drafts.length === 0) return [];
+	const db = getDatabase();
+	const maxSeq = db
+		.query("SELECT COALESCE(MAX(seq), -1) as max_seq FROM chunks WHERE tab_id = $tabId")
+		.get({ $tabId: tabId }) as { max_seq: number };
+	let seq = (maxSeq?.max_seq ?? -1) + 1;
+	const now = Date.now();
+	const insert = db.query(
+		`INSERT INTO chunks (id, tab_id, seq, turn_id, step, role, type, data_json, created_at)
+		 VALUES ($id, $tabId, $seq, $turnId, $step, $role, $type, $dataJson, $now)`,
+	);
+	const out: ChunkRow[] = [];
+	for (const draft of drafts) {
+		const id = randomUUID();
+		insert.run({
+			$id: id,
+			$tabId: tabId,
+			$seq: seq,
+			$turnId: draft.turnId,
+			$step: draft.step,
+			$role: draft.role,
+			$type: draft.type,
+			$dataJson: JSON.stringify(draft.data),
+			$now: now,
+		});
+		out.push({
+			id,
+			tabId,
+			seq,
+			turnId: draft.turnId,
+			step: draft.step,
+			role: draft.role,
+			type: draft.type,
+			data: draft.data,
+			createdAt: now,
+		});
+		seq++;
+	}
+	return out;
+}
+
+/**
+ * Read chunk rows for a tab in `seq` order (ASC). Pagination mirrors the old
+ * message pagination but at chunk granularity:
+ *   - no options → all rows;
+ *   - `before` → rows with `seq < before`, most-recent-first then reversed;
+ *   - `limit` → most recent `limit` rows, reversed to ASC.
+ */
+export function getChunksForTab(
+	tabId: string,
+	options?: { limit?: number; before?: number },
+): ChunkRow[] {
+	const db = getDatabase();
+	if (!options) {
+		const rows = db
+			.query("SELECT * FROM chunks WHERE tab_id = $tabId ORDER BY seq ASC")
+			.all({ $tabId: tabId }) as Array<Record<string, unknown>>;
+		return rows.map(mapRow);
+	}
+	const { limit, before } = options;
+	if (before !== undefined) {
+		if (limit !== undefined) {
+			const rows = db
+				.query(
+					"SELECT * FROM chunks WHERE tab_id = $tabId AND seq < $before ORDER BY seq DESC LIMIT $limit",
+				)
+				.all({ $tabId: tabId, $before: before, $limit: limit }) as Array<Record<string, unknown>>;
+			return rows.map(mapRow).reverse();
+		}
+		const rows = db
+			.query("SELECT * FROM chunks WHERE tab_id = $tabId AND seq < $before ORDER BY seq DESC")
+			.all({ $tabId: tabId, $before: before }) as Array<Record<string, unknown>>;
+		return rows.map(mapRow).reverse();
+	}
+	if (limit !== undefined) {
+		const rows = db
+			.query("SELECT * FROM chunks WHERE tab_id = $tabId ORDER BY seq DESC LIMIT $limit")
+			.all({ $tabId: tabId, $limit: limit }) as Array<Record<string, unknown>>;
+		return rows.map(mapRow).reverse();
+	}
+	const rows = db
+		.query("SELECT * FROM chunks WHERE tab_id = $tabId ORDER BY seq ASC")
+		.all({ $tabId: tabId }) as Array<Record<string, unknown>>;
+	return rows.map(mapRow);
+}
+
+/**
+ * Derived, grouped view of a tab's full history as messages. Used to
+ * pre-populate the agent's in-memory `ChatMessage[]` history when an Agent is
+ * (re)constructed. Always reads the full log (grouping a partial window would
+ * be lossy for the rebuild path).
+ */
+export function getMessagesForTab(tabId: string): MessageRow[] {
+	return groupRowsToMessages(getChunksForTab(tabId));
+}
+
+export function getTotalChunkCount(tabId: string): number {
+	const db = getDatabase();
+	const row = db
+		.query("SELECT COUNT(*) as count FROM chunks WHERE tab_id = $tabId")
+		.get({ $tabId: tabId }) as { count: number } | null;
+	return row?.count ?? 0;
+}
+
+export function clearChunksForTab(tabId: string): void {
+	const db = getDatabase();
+	db.query("DELETE FROM chunks WHERE tab_id = $tabId").run({ $tabId: tabId });
+}
diff --git a/packages/core/src/db/index.ts b/packages/core/src/db/index.ts
index e63b266..18dd1b5 100644
--- a/packages/core/src/db/index.ts
+++ b/packages/core/src/db/index.ts
@@ -93,16 +93,46 @@ export function getDatabase(): Database {
 		// Column already exists — ignore
 	}
 
-	_db.run(`CREATE TABLE IF NOT EXISTS messages (
-		id           TEXT PRIMARY KEY,
-		tab_id       TEXT NOT NULL REFERENCES tabs(id),
-		seq          INTEGER NOT NULL,
-		role         TEXT NOT NULL,
-		content_json TEXT NOT NULL,
-		created_at   INTEGER NOT NULL
+	// ─── Append-only chunk log (replaces the old `messages` blob table) ──
+	//
+	// A conversation is stored as a flat, append-only stream of chunk rows
+	// keyed by a per-tab monotonic `seq`. "Message" and "turn" are DERIVED
+	// groupings (see db/chunks.ts), never stored containers. This is what
+	// powers per-chunk frontend pagination AND the stable per-step wire
+	// format that fixes Anthropic prompt-cache churn (see plan-chunk-log.md).
+	//
+	//   role  : 'user' | 'assistant' | 'tool' | 'system'
+	//   type  : 'text' | 'thinking' | 'tool_call' | 'tool_result' | 'error' | 'system'
+	//   step  : LLM round-trip index within a turn (user/system rows = 0)
+	//   data_json: the type-specific payload (see ChunkData in types)
+	_db.run(`CREATE TABLE IF NOT EXISTS chunks (
+		id          TEXT PRIMARY KEY,
+		tab_id      TEXT NOT NULL,
+		seq         INTEGER NOT NULL,
+		turn_id     TEXT NOT NULL,
+		step        INTEGER NOT NULL DEFAULT 0,
+		role        TEXT NOT NULL,
+		type        TEXT NOT NULL,
+		data_json   TEXT NOT NULL,
+		created_at  INTEGER NOT NULL
 	)`);
 
-	_db.run(`CREATE INDEX IF NOT EXISTS idx_messages_tab ON messages(tab_id, seq)`);
+	_db.run(`CREATE INDEX IF NOT EXISTS idx_chunks_tab_seq ON chunks(tab_id, seq)`);
+
+	// One-shot migration off the legacy `messages` blob model. Beta software,
+	// no backward compatibility: the old chat history is destroyed (tabs +
+	// messages), while settings / credentials / api_keys / usage_cache /
+	// wake_schedule are preserved. Detect the old schema by the presence of
+	// the `messages` table; once dropped, this branch never runs again.
+	const hasLegacyMessages = _db
+		.query("SELECT name FROM sqlite_master WHERE type='table' AND name='messages'")
+		.get() as { name: string } | null;
+	if (hasLegacyMessages) {
+		_db.run("DROP TABLE IF EXISTS messages");
+		// Clear conversation containers too (fresh slate for the new model).
+		_db.run("DELETE FROM tabs");
+		_db.run("DELETE FROM chunks");
+	}
 
 	_db.run(`CREATE TABLE IF NOT EXISTS settings (
 		key   TEXT PRIMARY KEY,
diff --git a/packages/core/src/db/messages.ts b/packages/core/src/db/messages.ts
deleted file mode 100644
index 7fc6ccf..0000000
--- a/packages/core/src/db/messages.ts
+++ /dev/null
@@ -1,154 +0,0 @@
-import type { Chunk, MessageRole } from "../types/index.js";
-import { getDatabase } from "./index.js";
-
-/**
- * A persisted message row, with `content_json` already parsed into a `Chunk[]`.
- * Mirrors the new schema (no `thinking` column — that lived under the old
- * `content + toolCalls + toolResults + thinking` model).
- */
-export interface MessageRow {
-	id: string;
-	tabId: string;
-	seq: number;
-	role: MessageRole;
-	chunks: Chunk[];
-	createdAt: number;
-}
-
-/**
- * Append a new message to the tab. Caller passes the already-serialized
- * chunk list as `contentJson` (i.e. `JSON.stringify(chunks)`).
- */
-export function appendMessage(
-	tabId: string,
-	id: string,
-	role: MessageRole,
-	contentJson: string,
-): void {
-	const db = getDatabase();
-	const maxSeq = db
-		.query("SELECT COALESCE(MAX(seq), -1) as max_seq FROM messages WHERE tab_id = $tabId")
-		.get({ $tabId: tabId }) as { max_seq: number };
-	const seq = (maxSeq?.max_seq ?? -1) + 1;
-	db.query(
-		`INSERT INTO messages (id, tab_id, seq, role, content_json, created_at)
-		 VALUES ($id, $tabId, $seq, $role, $contentJson, $now)`,
-	).run({
-		$id: id,
-		$tabId: tabId,
-		$seq: seq,
-		$role: role,
-		$contentJson: contentJson,
-		$now: Date.now(),
-	});
-}
-
-/**
- * Replace the persisted chunks for an existing message. `contentJson` is
- * the already-serialized chunk list.
- */
-export function updateMessage(id: string, contentJson: string): void {
-	const db = getDatabase();
-	db.query("UPDATE messages SET content_json = $contentJson WHERE id = $id").run({
-		$id: id,
-		$contentJson: contentJson,
-	});
-}
-
-/**
- * Read messages for a tab in seq order (ASC). `content_json` is parsed into
- * `Chunk[]` here so callers don't have to. If a row's JSON is malformed,
- * the message is returned with an empty chunk list rather than throwing.
- *
- * When `options` is omitted, returns ALL messages (backward compatible).
- *
- * When `options.before` is provided, returns messages with `seq < before`,
- * taking the most recent ones first (DESC) up to `options.limit`, then
- * reversing back to ASC before returning.
- *
- * When only `options.limit` is provided, returns the most recent `limit`
- * messages, reversed back to ASC.
- */
-export function getMessagesForTab(
-	tabId: string,
-	options?: { limit?: number; before?: number },
-): MessageRow[] {
-	const db = getDatabase();
-
-	const mapRow = (row: Record<string, unknown>): MessageRow => {
-		const rawJson = row.content_json as string;
-		let chunks: Chunk[];
-		try {
-			const parsed = JSON.parse(rawJson);
-			chunks = Array.isArray(parsed) ? (parsed as Chunk[]) : [];
-		} catch {
-			chunks = [];
-		}
-		return {
-			id: row.id as string,
-			tabId: row.tab_id as string,
-			seq: row.seq as number,
-			role: row.role as MessageRole,
-			chunks,
-			createdAt: row.created_at as number,
-		};
-	};
-
-	// Backward-compatible path: no options → ALL messages, seq ASC.
-	if (!options) {
-		const rows = db
-			.query("SELECT * FROM messages WHERE tab_id = $tabId ORDER BY seq ASC")
-			.all({ $tabId: tabId }) as Array<Record<string, unknown>>;
-		return rows.map(mapRow);
-	}
-
-	const { limit, before } = options;
-
-	// Paginated path: fetch DESC, then reverse to ASC before returning.
-	if (before !== undefined) {
-		// `seq < before`, DESC, optionally limited.
-		if (limit !== undefined) {
-			const rows = db
-				.query(
-					"SELECT * FROM messages WHERE tab_id = $tabId AND seq < $before ORDER BY seq DESC LIMIT $limit",
-				)
-				.all({ $tabId: tabId, $before: before, $limit: limit }) as Array<Record<string, unknown>>;
-			return rows.map(mapRow).reverse();
-		}
-		const rows = db
-			.query("SELECT * FROM messages WHERE tab_id = $tabId AND seq < $before ORDER BY seq DESC")
-			.all({ $tabId: tabId, $before: before }) as Array<Record<string, unknown>>;
-		return rows.map(mapRow).reverse();
-	}
-
-	// Only `limit` provided: most recent `limit`, reversed to ASC.
-	if (limit !== undefined) {
-		const rows = db
-			.query("SELECT * FROM messages WHERE tab_id = $tabId ORDER BY seq DESC LIMIT $limit")
-			.all({ $tabId: tabId, $limit: limit }) as Array<Record<string, unknown>>;
-		return rows.map(mapRow).reverse();
-	}
-
-	// `options` was provided but empty → same as no options.
-	const rows = db
-		.query("SELECT * FROM messages WHERE tab_id = $tabId ORDER BY seq ASC")
-		.all({ $tabId: tabId }) as Array<Record<string, unknown>>;
-	return rows.map(mapRow);
-}
-
-/**
- * Return the total number of persisted messages for a tab.
- * Used by the API to advertise total history size alongside a paginated window.
- */
-export function getTotalMessageCount(tabId: string): number {
-	const db = getDatabase();
-	const row = db
-		.query("SELECT COUNT(*) as count FROM messages WHERE tab_id = $tabId")
-		.get({ $tabId: tabId }) as { count: number } | null;
-	return row?.count ?? 0;
-}
-
-export function clearMessagesForTab(tabId: string): void {
-	const db = getDatabase();
-	db.query("DELETE FROM messages WHERE tab_id = $tabId").run({ $tabId: tabId });
-}
diff --git a/packages/core/src/index.ts b/packages/core/src/index.ts
index e33ad2f..9fe7550 100644
--- a/packages/core/src/index.ts
+++ b/packages/core/src/index.ts
@@ -29,16 +29,19 @@ export {
 } from "./config/index.js";
 // Credentials
 export * from "./credentials/index.js";
-// Database
-export { closeDatabase, getDatabase, getDatabasePath } from "./db/index.js";
 export {
-	appendMessage,
-	clearMessagesForTab,
+	appendChunks,
+	clearChunksForTab,
+	explodeTurn,
+	explodeUserText,
+	getChunksForTab,
 	getMessagesForTab,
-	getTotalMessageCount,
+	getTotalChunkCount,
+	groupRowsToMessages,
 	type MessageRow,
-	updateMessage,
-} from "./db/messages.js";
+} from "./db/chunks.js";
+// Database
+export { closeDatabase, getDatabase, getDatabasePath } from "./db/index.js";
 export { deleteSetting, getSetting, setSetting } from "./db/settings.js";
 // Tabs & Messages
 export {
diff --git a/packages/core/src/types/index.ts b/packages/core/src/types/index.ts
index 7e7460b..3fcdb40 100644
--- a/packages/core/src/types/index.ts
+++ b/packages/core/src/types/index.ts
@@ -78,6 +78,85 @@ export interface ChatMessage {
 	chunks: Chunk[];
 }
 
+// ─── Append-only chunk log (persisted model) ─────────────────────
+//
+// The DB stores a conversation as a flat stream of `ChunkRow`s (see
+// db/chunks.ts). The render-facing `Chunk`/`ChatMessage` shapes above are
+// DERIVED from these rows by grouping (turn_id + step + role). Tool calls
+// and their results are SEPARATE rows linked by `callId`, mapping 1:1 to the
+// Anthropic wire format.
+
+/** Role of a persisted chunk row. `tool` rows hold tool results. */
+export type ChunkRole = "user" | "assistant" | "tool" | "system";
+
+/** Discriminator for a persisted chunk row's payload. */
+export type ChunkType = "text" | "thinking" | "tool_call" | "tool_result" | "error" | "system";
+
+export interface TextData {
+	text: string;
+}
+export interface ThinkingData {
+	text: string;
+	metadata?: Record<string, unknown>;
+}
+export interface ToolCallData {
+	callId: string;
+	name: string;
+	arguments: Record<string, unknown>;
+}
+export interface ToolResultData {
+	callId: string;
+	name: string;
+	result: string;
+	isError: boolean;
+	shellOutput?: { stdout: string; stderr: string };
+}
+export interface ErrorData {
+	message: string;
+	statusCode?: number;
+}
+export interface SystemData {
+	kind: SystemChunkKind;
+	text: string;
+}
+
+export type ChunkData =
+	| TextData
+	| ThinkingData
+	| ToolCallData
+	| ToolResultData
+	| ErrorData
+	| SystemData;
+
+/**
+ * A persisted chunk row — the append-only unit of conversation storage and
+ * the unit of frontend pagination. `seq` is per-tab monotonic and is both the
+ * ordering key and the pagination cursor.
+ */
+export interface ChunkRow {
+	id: string;
+	tabId: string;
+	seq: number;
+	turnId: string;
+	step: number;
+	role: ChunkRole;
+	type: ChunkType;
+	data: ChunkData;
+	createdAt: number;
+}
+
+/**
+ * A chunk-row draft (no `seq`/`tabId`/`createdAt`/`id` yet) used when
+ * exploding an in-memory turn into rows for persistence.
+ */
+export interface ChunkRowDraft {
+	turnId: string;
+	step: number;
+	role: ChunkRole;
+	type: ChunkType;
+	data: ChunkData;
+}
+
 export interface ToolCall {
 	id: string;
 	name: string;