Store file lists in BranchSummaryEntry.details for cumulative tracking

- BranchSummaryResult now returns readFiles and modifiedFiles separately
- BranchSummaryDetails type for details: { readFiles, modifiedFiles }
- branchWithSummary accepts optional details parameter
- Collect files from existing branch_summary.details when preparing entries
- Files accumulate across nested branch summaries

packages/ai/src/models.generated.ts

@@ -3835,13 +3835,13 @@ export const MODELS = {
 			reasoning: true,
 			input: ["text"],
 			cost: {
-				input: 0.39,
+				input: 0.35,
-				output: 1.9,
+				output: 1.5,
 				cacheRead: 0,
 				cacheWrite: 0,
 			},
-			contextWindow: 204800,
+			contextWindow: 202752,
-			maxTokens: 204800,
+			maxTokens: 65536,
 		} satisfies Model<"openai-completions">,
 		"z-ai/glm-4.6:exacto": {
 			id: "z-ai/glm-4.6:exacto",
@@ -6104,9 +6104,9 @@ export const MODELS = {
 			contextWindow: 32768,
 			maxTokens: 4096,
 		} satisfies Model<"openai-completions">,
-		"anthropic/claude-3.5-haiku": {
+		"anthropic/claude-3.5-haiku-20241022": {
-			id: "anthropic/claude-3.5-haiku",
+			id: "anthropic/claude-3.5-haiku-20241022",
-			name: "Anthropic: Claude 3.5 Haiku",
+			name: "Anthropic: Claude 3.5 Haiku (2024-10-22)",
 			api: "openai-completions",
 			provider: "openrouter",
 			baseUrl: "https://openrouter.ai/api/v1",
@@ -6121,9 +6121,9 @@ export const MODELS = {
 			contextWindow: 200000,
 			maxTokens: 8192,
 		} satisfies Model<"openai-completions">,
-		"anthropic/claude-3.5-haiku-20241022": {
+		"anthropic/claude-3.5-haiku": {
-			id: "anthropic/claude-3.5-haiku-20241022",
+			id: "anthropic/claude-3.5-haiku",
-			name: "Anthropic: Claude 3.5 Haiku (2024-10-22)",
+			name: "Anthropic: Claude 3.5 Haiku",
 			api: "openai-completions",
 			provider: "openrouter",
 			baseUrl: "https://openrouter.ai/api/v1",
@@ -6359,6 +6359,23 @@ export const MODELS = {
 			contextWindow: 128000,
 			maxTokens: 16384,
 		} satisfies Model<"openai-completions">,
+		"meta-llama/llama-3.1-8b-instruct": {
+			id: "meta-llama/llama-3.1-8b-instruct",
+			name: "Meta: Llama 3.1 8B Instruct",
+			api: "openai-completions",
+			provider: "openrouter",
+			baseUrl: "https://openrouter.ai/api/v1",
+			reasoning: false,
+			input: ["text"],
+			cost: {
+				input: 0.02,
+				output: 0.03,
+				cacheRead: 0,
+				cacheWrite: 0,
+			},
+			contextWindow: 131072,
+			maxTokens: 16384,
+		} satisfies Model<"openai-completions">,
 		"meta-llama/llama-3.1-405b-instruct": {
 			id: "meta-llama/llama-3.1-405b-instruct",
 			name: "Meta: Llama 3.1 405B Instruct",
@@ -6393,23 +6410,6 @@ export const MODELS = {
 			contextWindow: 131072,
 			maxTokens: 4096,
 		} satisfies Model<"openai-completions">,
-		"meta-llama/llama-3.1-8b-instruct": {
-			id: "meta-llama/llama-3.1-8b-instruct",
-			name: "Meta: Llama 3.1 8B Instruct",
-			api: "openai-completions",
-			provider: "openrouter",
-			baseUrl: "https://openrouter.ai/api/v1",
-			reasoning: false,
-			input: ["text"],
-			cost: {
-				input: 0.02,
-				output: 0.03,
-				cacheRead: 0,
-				cacheWrite: 0,
-			},
-			contextWindow: 131072,
-			maxTokens: 16384,
-		} satisfies Model<"openai-completions">,
 		"mistralai/mistral-nemo": {
 			id: "mistralai/mistral-nemo",
 			name: "Mistral: Mistral Nemo",
@@ -6546,6 +6546,23 @@ export const MODELS = {
 			contextWindow: 128000,
 			maxTokens: 4096,
 		} satisfies Model<"openai-completions">,
+		"openai/gpt-4o-2024-05-13": {
+			id: "openai/gpt-4o-2024-05-13",
+			name: "OpenAI: GPT-4o (2024-05-13)",
+			api: "openai-completions",
+			provider: "openrouter",
+			baseUrl: "https://openrouter.ai/api/v1",
+			reasoning: false,
+			input: ["text", "image"],
+			cost: {
+				input: 5,
+				output: 15,
+				cacheRead: 0,
+				cacheWrite: 0,
+			},
+			contextWindow: 128000,
+			maxTokens: 4096,
+		} satisfies Model<"openai-completions">,
 		"openai/gpt-4o": {
 			id: "openai/gpt-4o",
 			name: "OpenAI: GPT-4o",
@@ -6580,23 +6597,6 @@ export const MODELS = {
 			contextWindow: 128000,
 			maxTokens: 64000,
 		} satisfies Model<"openai-completions">,
-		"openai/gpt-4o-2024-05-13": {
-			id: "openai/gpt-4o-2024-05-13",
-			name: "OpenAI: GPT-4o (2024-05-13)",
-			api: "openai-completions",
-			provider: "openrouter",
-			baseUrl: "https://openrouter.ai/api/v1",
-			reasoning: false,
-			input: ["text", "image"],
-			cost: {
-				input: 5,
-				output: 15,
-				cacheRead: 0,
-				cacheWrite: 0,
-			},
-			contextWindow: 128000,
-			maxTokens: 4096,
-		} satisfies Model<"openai-completions">,
 		"meta-llama/llama-3-70b-instruct": {
 			id: "meta-llama/llama-3-70b-instruct",
 			name: "Meta: Llama 3 70B Instruct",
@@ -6835,23 +6835,6 @@ export const MODELS = {
 			contextWindow: 8191,
 			maxTokens: 4096,
 		} satisfies Model<"openai-completions">,
-		"openai/gpt-3.5-turbo": {
-			id: "openai/gpt-3.5-turbo",
-			name: "OpenAI: GPT-3.5 Turbo",
-			api: "openai-completions",
-			provider: "openrouter",
-			baseUrl: "https://openrouter.ai/api/v1",
-			reasoning: false,
-			input: ["text"],
-			cost: {
-				input: 0.5,
-				output: 1.5,
-				cacheRead: 0,
-				cacheWrite: 0,
-			},
-			contextWindow: 16385,
-			maxTokens: 4096,
-		} satisfies Model<"openai-completions">,
 		"openai/gpt-4": {
 			id: "openai/gpt-4",
 			name: "OpenAI: GPT-4",
@@ -6869,6 +6852,23 @@ export const MODELS = {
 			contextWindow: 8191,
 			maxTokens: 4096,
 		} satisfies Model<"openai-completions">,
+		"openai/gpt-3.5-turbo": {
+			id: "openai/gpt-3.5-turbo",
+			name: "OpenAI: GPT-3.5 Turbo",
+			api: "openai-completions",
+			provider: "openrouter",
+			baseUrl: "https://openrouter.ai/api/v1",
+			reasoning: false,
+			input: ["text"],
+			cost: {
+				input: 0.5,
+				output: 1.5,
+				cacheRead: 0,
+				cacheWrite: 0,
+			},
+			contextWindow: 16385,
+			maxTokens: 4096,
+		} satisfies Model<"openai-completions">,
 		"openrouter/auto": {
 			id: "openrouter/auto",
 			name: "OpenRouter: Auto Router",

packages/coding-agent/src/core/agent-session.ts

@@ -1644,6 +1644,7 @@ export class AgentSession {
 
 		// Run default summarizer if needed
 		let summaryText: string | undefined;
+		let summaryDetails: unknown;
 		if (options.summarize && entriesToSummarize.length > 0 && !hookSummary) {
 			const model = this.model!;
 			const apiKey = await this._modelRegistry.getApiKey(model);
@@ -1666,8 +1667,13 @@ export class AgentSession {
 				throw new Error(result.error);
 			}
 			summaryText = result.summary;
+			summaryDetails = {
+				readFiles: result.readFiles || [],
+				modifiedFiles: result.modifiedFiles || [],
+			};
 		} else if (hookSummary) {
 			summaryText = hookSummary.summary;
+			summaryDetails = hookSummary.details;
 		}
 
 		// Determine the new leaf position based on target type
@@ -1698,7 +1704,7 @@ export class AgentSession {
 		let summaryEntry: BranchSummaryEntry | undefined;
 		if (summaryText) {
 			// Create summary at target position (can be null for root)
-			const summaryId = this.sessionManager.branchWithSummary(newLeafId, summaryText);
+			const summaryId = this.sessionManager.branchWithSummary(newLeafId, summaryText, summaryDetails);
 			summaryEntry = this.sessionManager.getEntry(summaryId) as BranchSummaryEntry;
 		} else if (newLeafId === null) {
 			// No summary, navigating to root - reset leaf

packages/coding-agent/src/core/compaction/branch-summarization.ts

@@ -18,10 +18,18 @@ import { estimateTokens } from "./compaction.js";
 
 export interface BranchSummaryResult {
 	summary?: string;
+	readFiles?: string[];
+	modifiedFiles?: string[];
 	aborted?: boolean;
 	error?: string;
 }
 
+/** Details stored in BranchSummaryEntry.details for file tracking */
+export interface BranchSummaryDetails {
+	readFiles: string[];
+	modifiedFiles: string[];
+}
+
 export interface FileOperations {
 	read: Set<string>;
 	written: Set<string>;
@@ -183,6 +191,10 @@ function extractFileOpsFromMessage(message: AgentMessage, fileOps: FileOperation
  * Walks entries from NEWEST to OLDEST, adding messages until we hit the token budget.
  * This ensures we keep the most recent context when the branch is too long.
  *
+ * Also collects file operations from:
+ * - Tool calls in assistant messages
+ * - Existing branch_summary entries' details (for cumulative tracking)
+ *
  * @param entries - Entries in chronological order
  * @param tokenBudget - Maximum tokens to include (0 = no limit)
  */
@@ -195,13 +207,30 @@ export function prepareBranchEntries(entries: SessionEntry[], tokenBudget: numbe
 	};
 	let totalTokens = 0;
 
-	// Walk from newest to oldest to prioritize recent context
+	// First pass: collect file ops from ALL entries (even if they don't fit in token budget)
+	// This ensures we capture cumulative file tracking from nested branch summaries
+	for (const entry of entries) {
+		if (entry.type === "branch_summary" && entry.details) {
+			const details = entry.details as BranchSummaryDetails;
+			if (Array.isArray(details.readFiles)) {
+				for (const f of details.readFiles) fileOps.read.add(f);
+			}
+			if (Array.isArray(details.modifiedFiles)) {
+				// Modified files go into both edited and written for proper deduplication
+				for (const f of details.modifiedFiles) {
+					fileOps.edited.add(f);
+				}
+			}
+		}
+	}
+
+	// Second pass: walk from newest to oldest, adding messages until token budget
 	for (let i = entries.length - 1; i >= 0; i--) {
 		const entry = entries[i];
 		const message = getMessageFromEntry(entry);
 		if (!message) continue;
 
-		// Extract file ops from assistant messages
+		// Extract file ops from assistant messages (tool calls)
 		extractFileOpsFromMessage(message, fileOps);
 
 		const tokens = estimateTokens(message);
@@ -238,32 +267,6 @@ const BRANCH_SUMMARY_PROMPT = `Summarize this conversation branch concisely for
 
 Be brief and focused on what matters for future reference.`;
 
-/**
- * Format file operations as a static section to append to summary.
- */
-function formatFileOperations(fileOps: FileOperations): string {
-	// Combine edited and written into "modified"
-	const modified = new Set([...fileOps.edited, ...fileOps.written]);
-
-	// Read-only = read but not modified
-	const readOnly = [...fileOps.read].filter((f) => !modified.has(f)).sort();
-
-	const sections: string[] = [];
-
-	if (readOnly.length > 0) {
-		sections.push(`<read-files>\n${readOnly.join("\n")}\n</read-files>`);
-	}
-
-	if (modified.size > 0) {
-		const files = [...modified].sort();
-		sections.push(`<modified-files>\n${files.join("\n")}\n</modified-files>`);
-	}
-
-	if (sections.length === 0) return "";
-
-	return `\n\n${sections.join("\n\n")}`;
-}
-
 /**
  * Convert messages to text for the summarization prompt.
  */
@@ -358,13 +361,19 @@ export async function generateBranchSummary(
 		return { error: response.errorMessage || "Summarization failed" };
 	}
 
-	let summary = response.content
+	const summary = response.content
 		.filter((c): c is { type: "text"; text: string } => c.type === "text")
 		.map((c) => c.text)
 		.join("\n");
 
-	// Append static file operations section
+	// Compute file lists for details
-	summary += formatFileOperations(fileOps);
+	const modified = new Set([...fileOps.edited, ...fileOps.written]);
+	const readOnly = [...fileOps.read].filter((f) => !modified.has(f)).sort();
+	const modifiedFiles = [...modified].sort();
 
-	return { summary: summary || "No summary generated" };
+	return {
+		summary: summary || "No summary generated",
+		readFiles: readOnly,
+		modifiedFiles,
+	};
 }

packages/coding-agent/src/core/session-manager.ts

@@ -872,7 +872,7 @@ export class SessionManager {
 	 * Same as branch(), but also appends a branch_summary entry that captures
 	 * context from the abandoned conversation path.
 	 */
-	branchWithSummary(branchFromId: string | null, summary: string): string {
+	branchWithSummary(branchFromId: string | null, summary: string, details?: unknown): string {
 		if (branchFromId !== null && !this.byId.has(branchFromId)) {
 			throw new Error(`Entry ${branchFromId} not found`);
 		}
@@ -884,6 +884,7 @@ export class SessionManager {
 			timestamp: new Date().toISOString(),
 			fromId: branchFromId ?? "root",
 			summary,
+			details,
 		};
 		this._appendEntry(entry);
 		return entry.id;