@inixiative/archive 0.12.1 → 0.13.0

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@inixiative/archive",
-  "version": "0.12.1",
+  "version": "0.13.0",
   "type": "module",
   "exports": {
     ".": "./src/index.ts",
@@ -43,7 +43,7 @@
   },
   "devDependencies": {
     "@biomejs/biome": "2.5.2",
-    "@inixiative/config": "^0.5.66",
+    "@inixiative/config": "^0.5.70",
     "@types/bun": "1.4.2",
     "lefthook": "2.1.9",
     "typescript": "6.0.3"

package/src/import.ts

@@ -27,7 +27,35 @@ const injectedTags = [
   'user_instructions',
   'permissions',
   'INSTRUCTIONS',
+  'codex_internal_context',
+  'heartbeat',
+  'in-app-browser-context',
+  'external_codex_apps_open_page',
+  'send_user_message_question_reply',
+  'turn_aborted',
 ].join('|');
+/** Prompts Codex writes in the user's role. */
+const injectedPreamble = /^(?:# In app browser:|The following is the Codex agent history)/;
+/**
+ * Nothing but blocks of harness-style tags such as `<codex_internal_context>`: `snake_case` or
+ * `kebab-case` names, unlike markup a person pastes. A linear scan, since a backtracking pattern is
+ * exponential on many sibling blocks.
+ */
+function onlyTagBlocks(text: string) {
+  const open = /\s*<([a-z][a-z0-9]*(?:[_-][a-z0-9]+)+)(?:\s[^>]*)?>/y;
+  let at = 0;
+  let blocks = 0;
+  while (at < text.length) {
+    open.lastIndex = at;
+    const match = open.exec(text);
+    if (!match) return blocks > 0 && !text.slice(at).trim();
+    const close = text.indexOf(`</${match[1]}>`, open.lastIndex);
+    if (close < 0) return false;
+    at = close + match[1]!.length + 3;
+    blocks++;
+  }
+  return blocks > 0;
+}
 const injectedBlock = new RegExp(`<(${injectedTags})(?:\\s[^>]*)?>[\\s\\S]*?</\\1>`, 'g');
 const injectedOpening = new RegExp(`^<(?:${injectedTags})(?:>|\\s)`);
 /** Foundry's native-session envelope; the person's text is the first User Message section. */
@@ -35,7 +63,7 @@ const foundryUserMessage =
   /^# User Message[ \t]*\n([\s\S]*?)(?=\n# (?:User Message|Assistant Message|Foundry Injection)[ \t]*\n|(?![\s\S]))/m;
 
 /** What the person typed in a user message, without injected content; undefined when nothing is left. */
-function typedPrompt(text: string) {
+export function typedPrompt(text: string) {
   let prompt = text
     .replace(injectedBlock, '')
     .replace(/<\/?pasted_content(?:\s[^>]*)?>/g, '')
@@ -43,7 +71,13 @@ function typedPrompt(text: string) {
     .trim();
   if (/^# (?:System Context|User Message)[ \t]*\n/.test(prompt))
     prompt = prompt.match(foundryUserMessage)?.[1].trim() ?? '';
-  if (!prompt || injectedOpening.test(prompt) || /^\[Request interrupted by user/.test(prompt))
+  if (
+    !prompt ||
+    injectedOpening.test(prompt) ||
+    injectedPreamble.test(prompt) ||
+    onlyTagBlocks(prompt) ||
+    /^\[Request interrupted by user/.test(prompt)
+  )
     return undefined;
   return prompt;
 }
@@ -66,6 +100,8 @@ export function importTranscriptLines(
   let contentBytes = 0;
   // Codex states the model per turn; Claude Code per assistant record.
   let turn: { model?: string; effort?: string } = {};
+  // A Codex session started by a script (codex exec, a custom originator) has no typed prompts.
+  let scripted = false;
   for (const rawLine of lines) {
     index++;
     const line = rawLine.trim();
@@ -83,6 +119,12 @@ export function importTranscriptLines(
       if (sessionId && id && id !== sessionId)
         throw new Error('Transcript session identity mismatch');
       sessionId ??= id;
+      // Subagent and guardian threads, and `codex exec`, are driven by an agent or a script.
+      const { originator, source } = row.payload ?? {};
+      if (typeof originator === 'string')
+        scripted = !/^codex/i.test(originator) || /^codex_(?:exec|sdk|mcp)/.test(originator);
+      if (source === 'exec' || (source && typeof source === 'object' && 'subagent' in source))
+        scripted = true;
     }
     if (options.source === 'claude-code' && row.sessionId) {
       // Subagent transcripts reuse their parent's sessionId; each is its own archive.
@@ -102,6 +144,26 @@ export function importTranscriptLines(
         : row.type === 'assistant'
           ? { model: text(row.message?.model), effort: text(row.effort) }
           : {};
+    // A user record the person did not type: an expansion, a compaction summary, a subagent's prompt,
+    // a message from the harness, another agent or an SDK caller, or a scripted Codex session's prompt.
+    // It is kept as an event, not a user turn.
+    const written =
+      options.source === 'claude-code'
+        ? row.type === 'user' &&
+          Boolean(
+            row.isMeta ||
+              row.isCompactSummary ||
+              row.isSidechain ||
+              (row.origin && typeof row.origin === 'object'
+                ? row.origin.kind !== 'human'
+                : typeof row.entrypoint === 'string' && row.entrypoint.startsWith('sdk')),
+          )
+        : scripted && row.type === 'response_item' && row.payload?.role === 'user';
+    // User text is a user entry only when the person typed it; markers and injected blocks are events.
+    const textKind = (role: 'user' | 'assistant', content: unknown) =>
+      role === 'user' && (written || typeof content !== 'string' || !typedPrompt(content))
+        ? 'event'
+        : role;
     const timestamp = typeof row.timestamp === 'number' ? row.timestamp : Date.parse(row.timestamp);
     const add = (kind: ArchiveEntry['kind'], content: unknown, suffix: string, callId?: string) => {
       if (typeof content !== 'string' || !content.length) return;
@@ -118,10 +180,10 @@ export function importTranscriptLines(
           'Supported transcript content exceeds the archive size limit; no archive imported',
         );
       seen.set(id, identity);
-      // Claude Code marks expanded commands and caveats as meta: a title only when nothing was typed.
-      if (kind === 'user' && !prompt && !row.isCompactSummary) {
-        if (!row.isMeta) prompt = typedPrompt(content);
-        else metaPrompt ??= typedPrompt(content);
+      // A prompt someone else wrote (an expansion, a subagent's task) titles the session only when nothing was typed.
+      if (!prompt && !row.isCompactSummary) {
+        if (kind === 'user') prompt = typedPrompt(content);
+        else if (kind === 'event' && written) metaPrompt ??= typedPrompt(content);
       }
       entries.push({
         id,
@@ -139,12 +201,16 @@ export function importTranscriptLines(
       });
     };
     if (options.source === 'codex') {
-      if (row.type === 'response_item') {
+      // Codex records the person stopping a turn as its own event; Claude Code writes this marker.
+      if (row.type === 'event_msg' && row.payload?.type === 'turn_aborted') {
+        if (row.payload.reason === 'interrupted')
+          add('event', '[Request interrupted by user]', 'aborted');
+      } else if (row.type === 'response_item') {
         const p = row.payload ?? {};
         if (p.type === 'message' && ['user', 'assistant'].includes(p.role)) {
           for (const [i, block] of (p.content ?? []).entries()) {
             if (['input_text', 'output_text', 'text'].includes(block.type))
-              add(p.role, block.text, `text-${i}`);
+              add(textKind(p.role, block.text), block.text, `text-${i}`);
             else omissions.add('Non-text message content is not included');
           }
         } else if (['function_call', 'custom_tool_call'].includes(p.type)) {
@@ -181,10 +247,10 @@ export function importTranscriptLines(
       }
     } else if (['user', 'assistant'].includes(row.type) && row.message) {
       const blocks = row.message.content;
-      if (typeof blocks === 'string') add(row.type, blocks, 'text');
+      if (typeof blocks === 'string') add(textKind(row.type, blocks), blocks, 'text');
       else if (Array.isArray(blocks))
         for (const [i, block] of blocks.entries()) {
-          if (block.type === 'text') add(row.type, block.text, `text-${i}`);
+          if (block.type === 'text') add(textKind(row.type, block.text), block.text, `text-${i}`);
           else if (block.type === 'tool_use')
             add(
               'tool-call',