@inixiative/archive 0.2.0 → 0.2.1

package/README.md

@@ -98,3 +98,7 @@ bun run typecheck
 ```
 
 See [provenance](docs/PROVENANCE.md), [license](LICENSE), and [goals](tickets/README.md). The historical tickets describe earlier scope; this implementation now includes explicit provider collection and authenticated standalone hosting.
+
+### ChatGPT conversation text
+
+`archive import --file conversation.json --source chatgpt --project-id personal` accepts the JSON result of the desktop ChatGPT conversation reader (`thread`, `page`, and `turns`). This is a partial text capture, not a ChatGPT account export: tools, attachments, alternate branches and resolved citation targets are unavailable. ChatGPT conversations retain their own source identity.

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@inixiative/archive",
-  "version": "0.2.0",
+  "version": "0.2.1",
   "type": "module",
   "exports": {
     ".": "./src/index.ts",
@@ -12,7 +12,8 @@
     "./import-file": "./src/import-file.ts",
     "./preview": "./src/preview.ts",
     "./cli": "./src/cli.ts",
-    "./collector": "./src/collector.ts"
+    "./collector": "./src/collector.ts",
+    "./chatgpt": "./src/chatgpt.ts"
   },
   "files": [
     "src"
@@ -32,11 +33,11 @@
     "zod": "^4.3.6"
   },
   "devDependencies": {
-    "@types/bun": "1.4.2",
-    "typescript": "6.0.3",
     "@biomejs/biome": "2.5.2",
-    "@inixiative/config": "^0.5.33",
-    "lefthook": "2.1.9"
+    "@inixiative/config": "^0.5.34",
+    "@types/bun": "1.4.2",
+    "lefthook": "2.1.9",
+    "typescript": "6.0.3"
   },
   "bin": {
     "archive": "src/cli.ts"

package/src/chatgpt.ts

@@ -0,0 +1,74 @@
+import { z } from 'zod';
+import { type ArchiveEntry, archiveSnapshotSchema } from './index';
+
+const exportSchema = z.object({
+  thread: z.object({ id: z.string(), kind: z.literal('chatgpt'), title: z.string() }),
+  page: z.object({ order: z.literal('newest_first'), hasMore: z.boolean() }),
+  turns: z.array(
+    z.object({
+      id: z.string(),
+      startedAt: z.number().nullable(),
+      completedAt: z.number().nullable(),
+      items: z.array(
+        z.object({
+          type: z.string(),
+          id: z.string(),
+          text: z.string().optional(),
+          content: z.array(z.object({ type: z.string(), text: z.string().optional() })).optional(),
+        }),
+      ),
+    }),
+  ),
+});
+
+/** Import the text returned by the desktop ChatGPT conversation reader, not a native JSONL log. */
+export function importChatGPTThread(
+  input: unknown,
+  options: { sourceId: string; projectId?: string; title?: string },
+) {
+  const data = exportSchema.parse(input);
+  const entries: ArchiveEntry[] = [];
+  for (const turn of [...data.turns].reverse()) {
+    for (const item of turn.items) {
+      if (!['userMessage', 'agentMessage'].includes(item.type)) continue;
+      const text =
+        item.text ??
+        item.content
+          ?.filter((b) => b.type === 'text')
+          .map((b) => b.text ?? '')
+          .join('\n');
+      if (!text) continue;
+      entries.push({
+        id: item.id,
+        turnId: turn.id,
+        kind: item.type === 'userMessage' ? 'user' : 'assistant',
+        text,
+        timestamp:
+          (item.type === 'userMessage' ? turn.startedAt : turn.completedAt) === null
+            ? null
+            : (item.type === 'userMessage' ? turn.startedAt! : turn.completedAt!) * 1000,
+        sourceRef: `chatgpt:${data.thread.id}:${item.id}`,
+      });
+    }
+  }
+  if (!entries.length) throw new Error('ChatGPT export contains no readable messages');
+  return archiveSnapshotSchema.parse({
+    schemaVersion: 1,
+    source: 'chatgpt',
+    sourceId: options.sourceId,
+    sessionId: data.thread.id,
+    title: options.title ?? data.thread.title,
+    projectId: options.projectId,
+    tags: [],
+    capturedAt: Date.now(),
+    entries,
+    coverage: {
+      reasoning: 'unavailable',
+      completeness: 'partial',
+      omissions: [
+        'Desktop conversation text only; tool activity, attachments, alternate branches and citation targets are not exported. Reader text may be truncated.',
+        ...(data.page.hasMore ? ['Older conversation turns are not included in this page.'] : []),
+      ],
+    },
+  });
+}

package/src/cli.ts

@@ -4,6 +4,7 @@ import { chmodSync, existsSync, mkdirSync, readFileSync, writeFileSync } from 'n
 import { homedir } from 'node:os';
 import { join, resolve } from 'node:path';
 import { parseArgs } from 'node:util';
+import { importChatGPTThread } from './chatgpt';
 import {
   archiveRequest,
   publishArchive,
@@ -142,15 +143,28 @@ export async function runCli(args = Bun.argv.slice(2)) {
   const store = new LocalArchiveStore(storePath);
   try {
     if (command === 'import') {
-      if (!v.file || !['codex', 'claude-code'].includes(v.source ?? '') || !v['project-id'])
-        throw new Error('Import requires --file, --source codex|claude-code and --project-id');
-      const snapshot = importTranscriptFile(v.file, {
-        source: v.source as 'codex' | 'claude-code',
-        sourceId: store.sourceId,
-        projectId: v['project-id'],
-        sessionId: v['session-id'],
-        title: v.title,
-      });
+      if (
+        !v.file ||
+        !['codex', 'claude-code', 'chatgpt'].includes(v.source ?? '') ||
+        !v['project-id']
+      )
+        throw new Error(
+          'Import requires --file, --source codex|claude-code|chatgpt and --project-id',
+        );
+      const snapshot =
+        v.source === 'chatgpt'
+          ? importChatGPTThread(JSON.parse(readFileSync(v.file, 'utf8')), {
+              sourceId: store.sourceId,
+              projectId: v['project-id'],
+              title: v.title,
+            })
+          : importTranscriptFile(v.file, {
+              source: v.source as 'codex' | 'claude-code',
+              sourceId: store.sourceId,
+              projectId: v['project-id'],
+              sessionId: v['session-id'],
+              title: v.title,
+            });
       snapshot.tags = [
         ...new Set([...(store.read(archiveKey(snapshot))?.snapshot.tags ?? []), ...(v.tag ?? [])]),
       ];

package/src/index.ts

@@ -16,7 +16,7 @@ export const archiveSnapshotSchema = z
   .strictObject({
     schemaVersion: z.literal(1),
     sourceId: z.uuid(),
-    source: z.enum(['foundry', 'claude-code', 'codex']),
+    source: z.enum(['foundry', 'claude-code', 'codex', 'chatgpt']),
     sessionId: identifier,
     title: z.string().min(1).max(500),
     projectId: identifier.optional(),

package/src/server.ts

@@ -12,6 +12,7 @@ const ingestSchema = z.strictObject({
 });
 const searchSchema = z.strictObject({
   query: z.string().max(1000).default(''),
+  tag: z.string().min(1).max(120).optional(),
   budget: z.number().int().min(16).max(32768).default(2048),
   projectId: z.string().max(256).optional(),
   limit: z.number().int().min(1).max(100).default(20),
@@ -75,7 +76,11 @@ export function createArchiveHandler(store: LocalArchiveStore, token: string) {
         let remaining = input.budget;
         const matching = store
           .list()
-          .filter((a) => !input.projectId || a.projectId === input.projectId)
+          .filter(
+            (a) =>
+              (!input.projectId || a.projectId === input.projectId) &&
+              (!input.tag || a.tags.includes(input.tag)),
+          )
           .map((a) => ({
             a,
             selected: selectChunks(store.read(a.id)!.chunks, input.query, input.budget),