meetstream-migrate 2.3.0__tar.gz → 2.5.0__tar.gz

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (21) hide show
  1. meetstream_migrate-2.5.0/PKG-INFO +326 -0
  2. meetstream_migrate-2.5.0/README.md +290 -0
  3. {meetstream_migrate-2.3.0 → meetstream_migrate-2.5.0}/meetstream_migrate/__init__.py +1 -1
  4. {meetstream_migrate-2.3.0 → meetstream_migrate-2.5.0}/meetstream_migrate/cli.py +21 -4
  5. meetstream_migrate-2.5.0/meetstream_migrate/data/recall-v1-11.json +110 -0
  6. {meetstream_migrate-2.3.0 → meetstream_migrate-2.5.0}/meetstream_migrate/mappings.py +81 -32
  7. {meetstream_migrate-2.3.0 → meetstream_migrate-2.5.0}/meetstream_migrate/transformer.py +4 -1
  8. meetstream_migrate-2.5.0/meetstream_migrate.egg-info/PKG-INFO +326 -0
  9. {meetstream_migrate-2.3.0 → meetstream_migrate-2.5.0}/meetstream_migrate.egg-info/SOURCES.txt +3 -1
  10. {meetstream_migrate-2.3.0 → meetstream_migrate-2.5.0}/setup.py +3 -1
  11. meetstream_migrate-2.5.0/test/test_mappings.py +74 -0
  12. meetstream_migrate-2.3.0/PKG-INFO +0 -253
  13. meetstream_migrate-2.3.0/README.md +0 -217
  14. meetstream_migrate-2.3.0/meetstream_migrate.egg-info/PKG-INFO +0 -253
  15. {meetstream_migrate-2.3.0 → meetstream_migrate-2.5.0}/meetstream_migrate/scanner.py +0 -0
  16. {meetstream_migrate-2.3.0 → meetstream_migrate-2.5.0}/meetstream_migrate/telemetry.py +0 -0
  17. {meetstream_migrate-2.3.0 → meetstream_migrate-2.5.0}/meetstream_migrate.egg-info/dependency_links.txt +0 -0
  18. {meetstream_migrate-2.3.0 → meetstream_migrate-2.5.0}/meetstream_migrate.egg-info/entry_points.txt +0 -0
  19. {meetstream_migrate-2.3.0 → meetstream_migrate-2.5.0}/meetstream_migrate.egg-info/requires.txt +0 -0
  20. {meetstream_migrate-2.3.0 → meetstream_migrate-2.5.0}/meetstream_migrate.egg-info/top_level.txt +0 -0
  21. {meetstream_migrate-2.3.0 → meetstream_migrate-2.5.0}/setup.cfg +0 -0
@@ -0,0 +1,326 @@
1
+ Metadata-Version: 2.4
2
+ Name: meetstream-migrate
3
+ Version: 2.5.0
4
+ Summary: One-command migration from Recall.ai to MeetStream.ai — scans your codebase, maps APIs, transforms code.
5
+ Home-page: https://github.com/meetstream/recall-meetstream-migration-kit
6
+ Author: MeetStream
7
+ Author-email: sid@meetstream.ai
8
+ License: MIT
9
+ Keywords: meetstream,recall,migration,meeting-bot,api
10
+ Classifier: Development Status :: 4 - Beta
11
+ Classifier: Intended Audience :: Developers
12
+ Classifier: License :: OSI Approved :: MIT License
13
+ Classifier: Programming Language :: Python :: 3
14
+ Classifier: Programming Language :: Python :: 3.8
15
+ Classifier: Programming Language :: Python :: 3.9
16
+ Classifier: Programming Language :: Python :: 3.10
17
+ Classifier: Programming Language :: Python :: 3.11
18
+ Classifier: Programming Language :: Python :: 3.12
19
+ Classifier: Topic :: Software Development :: Libraries :: Python Modules
20
+ Requires-Python: >=3.8
21
+ Description-Content-Type: text/markdown
22
+ Requires-Dist: posthog
23
+ Requires-Dist: rich>=13.0.0
24
+ Requires-Dist: requests>=2.28.0
25
+ Dynamic: author
26
+ Dynamic: author-email
27
+ Dynamic: classifier
28
+ Dynamic: description
29
+ Dynamic: description-content-type
30
+ Dynamic: home-page
31
+ Dynamic: keywords
32
+ Dynamic: license
33
+ Dynamic: requires-dist
34
+ Dynamic: requires-python
35
+ Dynamic: summary
36
+
37
+ # meetstream-migrate
38
+
39
+ **One-command migration from [Recall.ai](https://recall.ai) to [MeetStream.ai](https://meetstream.ai).** Scans your codebase, maps every API call, shows the diff, and migrates — safely, with warnings for anything that needs a human look.
40
+
41
+ > **v2.5.0 — mappings for the MeetStream API (v1.10.0) and the Recall.ai API v1.11, November 2026.**
42
+ > MeetStream matches Recall across recording, per-participant streams, real-time audio/video/
43
+ > transcript, chat + image output, Google **and** Outlook calendar, scheduling, signed-in bots,
44
+ > and retention — plus voice agents in the meeting (Agents API) that Recall doesn't have.
45
+ > Pause/resume recording maps 1:1, Zoom bots need no setup (API key + meeting link), and live
46
+ > camera video into the call uses `bot_camera_stream` (WebRTC). See
47
+ > [Feature parity & gaps](#feature-parity--gaps) below.
48
+
49
+ ---
50
+
51
+ **Two ways to integrate MeetStream.** (1) The **Bot API**: meeting bots and recording, the direct replacement for Recall bots. (2) The **Agents API**: a voice agent (API name: MIA) on the bot, which talks in the meeting. This migrator moves Recall code to the Bot API; add an agent later with `agent_config_id` on `create_bot`. Guide: [Two ways to integrate](https://docs.meetstream.ai/guides/introduction/integration-paths).
52
+
53
+ ## Quick Start
54
+
55
+ ```bash
56
+ pip install meetstream-migrate
57
+ meetstream-migrate scan ./your-recall-project
58
+ ```
59
+
60
+ ## Commands
61
+
62
+ ```bash
63
+ # Scan (read-only) — finds every Recall.ai reference, writes NOTHING
64
+ meetstream-migrate scan ./your-project
65
+
66
+ # Dry run — shows exactly what would change, still writes nothing
67
+ meetstream-migrate migrate ./your-project --dry-run
68
+
69
+ # Migrate — applies all transformations (prompts before writing)
70
+ meetstream-migrate migrate ./your-project --yes
71
+
72
+ # List Recall.ai features with no MeetStream mapping
73
+ meetstream-migrate gaps
74
+
75
+ # Verify your MeetStream credentials + API reachability
76
+ meetstream-migrate test --api-key YOUR_MEETSTREAM_API_KEY
77
+ ```
78
+
79
+ ### CLI reference
80
+
81
+ ```
82
+ meetstream-migrate <command> [path] [options]
83
+
84
+ Commands:
85
+ scan [path] Scan only — show what would change
86
+ migrate [path] Scan & migrate a project
87
+ test Run API tests against MeetStream.ai
88
+ gaps List Recall.ai features with no MeetStream mapping
89
+
90
+ Options:
91
+ --yes, -y Skip the confirmation prompt
92
+ --dry-run Show changes without writing files
93
+ --api-key KEY MeetStream API key (for the test command)
94
+ --verbose, -v Detailed output
95
+ --help, -h Show help
96
+ ```
97
+
98
+ ## What it does
99
+
100
+ **1. Scan** — recursively finds:
101
+ - Recall.ai base URLs (`us-east-1.recall.ai`, `eu-west-1.recall.ai`, `api.recall.ai`)
102
+ - API endpoint calls (`/api/v1/bot/`, `/api/v1/bot/{id}/audio/`, etc.)
103
+ - Recall-specific field names (`meeting_url`, `recording_mode`, `transcription_options`, `metadata`, …)
104
+ - Webhook handlers (`bot.status_change`, `joining_call`, `in_call_recording`, `call_ended`, `done`, `fatal`, …)
105
+ - SDK imports (`require('recall-ai')`, `import recall`)
106
+ - Environment variables (`RECALL_API_KEY`, `RECALL_API_URL`, `RECALL_WEBHOOK_SECRET`)
107
+
108
+ Supported file types: `.js .ts .jsx .tsx .mjs .cjs .py .go .rb .java .kt .rs .php .cs .swift .env .yaml .yml .json .toml .ini .cfg`
109
+
110
+ **2. Display** — endpoint mapping table, field mapping table, webhook event mapping, and a compatibility check (auto-migratable vs. needs manual review).
111
+
112
+ **3. Transform** — base URL swap, endpoint path rewrites, field renames, webhook event renames, env var renames, import/require updates. Every change carries an inline warning when it needs a second look.
113
+
114
+ **4. Report** — what changed, what needs manual review, and a post-migration checklist.
115
+
116
+ ---
117
+
118
+ ## Endpoint Mapping
119
+
120
+ | Recall.ai | MeetStream.ai | Notes |
121
+ |-----------|---------------|-------|
122
+ | `POST /api/v1/bot/` | `POST /api/v1/bots/create_bot` | Field names differ (see below) |
123
+ | `GET /api/v1/bot/{id}/` | `GET /api/v1/bots/{id}/detail` | Also `GET /api/v1/bots/{id}/status` |
124
+ | `POST /api/v1/bot/{id}/leave_call/` | `GET /api/v1/bots/{id}/remove_bot` | ⚠️ Method changes POST → GET |
125
+ | `GET /api/v1/bot/` | `GET /api/v1/bots` | List all bots |
126
+ | `GET /bot/{id}/audio/` | `GET /bots/{id}/get_audio` | Download link, valid 1 hour |
127
+ | `GET /bot/{id}/video/` | `GET /bots/{id}/get_video` | Download link, valid 1 hour |
128
+ | `GET /bot/{id}/transcript/` | `GET /bots/{id}/get_transcript` | Takes the `bot_id`; for a specific transcript use `GET /transcript/{transcript_id}/get_transcript` |
129
+ | `GET /bot/{id}/participants/` | `GET /bots/{id}/get_participants` | |
130
+ | `GET /bot/{id}/chat_messages/` | `GET /bots/{id}/get_chats` | |
131
+ | `POST /bot/{id}/send_chat_message/` | `POST /bots/{id}/send_message` | REST body `{ "message": "..." }` — drop Recall's `to`/`pin` fields |
132
+ | `POST /bot/{id}/output_video/` | `POST /bots/{id}/send_image` | `img` (base64, from `b64_data`) or a public `img_url` |
133
+ | `POST /bot/{id}/output_audio/` | WebSocket `sendaudio` command | ⚠️ Manual — no REST equivalent, use `socket_connection_url` bridge |
134
+ | `POST /bot/{id}/output_media/` | `bot_camera_stream.websocket_url` on `create_bot` | ⚠️ Manual — publish WebRTC video through a `wss://` signaling socket; no webpage field |
135
+ | `POST /bot/{id}/pause_recording/` | `POST /bots/{id}/pause_recording` | ✅ 1:1 (empty body) |
136
+ | `POST /bot/{id}/resume_recording/` | `POST /bots/{id}/resume_recording` | ✅ 1:1 (empty body) |
137
+ | `GET /bot/{id}/screenshots/` | `GET /bots/{id}/screenshots` | |
138
+ | `GET /bot/{id}/speaker_timeline/` | `GET /bots/{id}/get_speaker_timeline` | Large timelines return `speaker_timeline_url` (download link, valid 1 h) instead of inline JSON |
139
+ | `POST /bot/{id}/delete_media/` | `DELETE /bots/{id}/delete` | ⚠️ Method POST → DELETE; permanent |
140
+ | `PATCH /api/v1/bot/{id}/` (scheduled bot) | `PATCH /api/v1/calendar/scheduled_bots/{bot_id}` | New `scheduled_join_time` and other create fields; only while `Scheduled` |
141
+ | `DELETE /api/v1/bot/{id}/` (scheduled bot) | `DELETE /api/v1/calendar/scheduled_bots/{bot_id}` | To delete media instead: `DELETE /bots/{id}/delete` |
142
+ | `POST /recording/{id}/create_transcript/` | `POST /bots/{bot_id}/transcribe` | Keyed by `bot_id`; list runs with `GET /bots/{bot_id}/transcriptions` (flagged) |
143
+ | `GET /recording/{id}/`, `/meeting_metadata/{id}/` | `GET /bots/{bot_id}/detail` | One bot = one recording (flagged) |
144
+ | `GET /transcript/{id}/` | `GET /transcript/{transcript_id}/get_transcript` | Returns segments; a failed transcript is `422` with the reason (flagged) |
145
+ | `GET /audio_mixed/`, `/video_mixed/` | `GET /bots/{bot_id}/get_audio`, `get_video` | flagged |
146
+ | `GET /audio_separate/`, `/video_separate/` | `GET /bots/{bot_id}/get_audio_streams`, `get_recording_streams` | Needs `audio_separate_streams` / `video_separate_streams` (flagged) |
147
+ | `GET /participant_events/` | `GET /bots/{bot_id}/get_participants`, `get_speaker_timeline` | flagged |
148
+ | `POST`/`DELETE /api/v2/calendar-events/{id}/bot/` | `POST`/`DELETE /api/v1/calendar/schedule/{event_id}` | `bot_config` fields go in the body; `deduplication_key` works |
149
+ | `GET /api/v2/calendar-events/` | `GET /api/v1/calendar/events` | |
150
+ | `/api/v2/calendars/` | `GET /api/v1/calendar/calendars`; create `POST /api/v1/calendar/create_calendar` or `create_outlook_calendar`; `DELETE /api/v1/calendar/disconnect` | ⚠️ Create and delete need manual edits |
151
+ | `POST /api/v1/meeting_direct_connect/` | `POST /api/v1/meeting_direct_connect` | Zoom (RTMS) only |
152
+ | `/api/v2/google-login-groups/`, `/api/v2/google-logins/` | `/api/v1/google-login-domains`, `/api/v1/google-logins` | ⚠️ Bodies differ (SSO key and certificate) |
153
+ | `start_recording`, `stop_recording`, `output_screenshare`, `pin_participant`, `billing/usage`, `zoom-oauth-*`, `sdk_upload` | none | Flagged by `scan`; see `gaps` |
154
+ | `POST /api/v1/webhook/` | Workspace webhooks (Dashboard > Configure > Webhooks, signed) or `callback_url` on `create_bot` (not signed) | No API call registers webhooks |
155
+ | `GET /bot/{id}/calendar/…` | `POST /calendar/create_calendar` (underscore) | Body: `google_client_id`/`google_client_secret`/`google_refresh_token` |
156
+
157
+ ## Field Mapping
158
+
159
+ | Recall.ai | MeetStream.ai | Notes |
160
+ |-----------|---------------|-------|
161
+ | `meeting_url` | `meeting_link` | |
162
+ | `recording_mode` (string) | `video_required` (boolean) | `"speaker_view"`/`"gallery_view"` → `true`, `"audio_only"` → `false` |
163
+ | `metadata` | `custom_attributes` | Values must be strings |
164
+ | `transcription_options` | `recording_config.transcript.provider.{provider}` | Object structure differs — see provider config |
165
+ | `noone_joined_timeout` | `no_one_joined_timeout` | |
166
+ | `recording_permission_denied_timeout` | same name in `automatic_leave` | Zoom only; 60-600 s, default 600 |
167
+ | `zoom.zak_url` / `zoom.obf_token_url` | `zoom.zak_url` / `zoom.obf_url` | Optional; uses your own Zoom app (`use_own_app: true`) |
168
+ | `zoom.join_token_url` | — | Remove: Zoom bots join with no setup |
169
+ | `destination_url` | `webhook_url` (nested in `live_transcription_required`) | |
170
+ | `assembly_ai` | `assemblyai` | Also: `assemblyai` uses `speech_models[]` + `language_code`, not `model`/`language` |
171
+ | `output_media.websocket.url` | `socket_connection_url.websocket_url` | |
172
+ | `output_media.camera` (webpage) | `bot_camera_stream.websocket_url` | ⚠️ Manual — you publish WebRTC video; MeetStream does not render a webpage |
173
+
174
+ ## Webhook Events — live-verified against production
175
+
176
+ MeetStream's envelope key is **`event`** — the **same key Recall uses**. No envelope rename needed. Alongside it: `bot_id`, `message`, `status_code` (200 success / 500 failure), `custom_attributes`, and on lifecycle events, `bot_status`.
177
+
178
+ ```json
179
+ { "event": "bot.inmeeting", "bot_id": "...", "bot_status": "InMeeting",
180
+ "message": "...", "status_code": 200, "custom_attributes": {} }
181
+ ```
182
+
183
+ **Two-layer model:** `bot.stopped` fires **once**; `bot_status` tells you *why* — `Stopped` (normal) / `NotAllowed` (lobby timeout) / `Denied` (host denied) / `Failed` (the meeting blocked the join) / `Error` (crash). `bot_event` gives the detail (`bot.kicked`, `bot.denied`, `bot.notallowed`, `bot.failed`). Recall's separate end-reason events collapse into this single event + status.
184
+
185
+ | Recall.ai Event | MeetStream `event` | `bot_status` | status_code |
186
+ |----------------|--------------------|--------------|-------------|
187
+ | `bot.joining_call` | `bot.joining` (may fire up to 3×) | `Joining` | 200 |
188
+ | `bot.in_waiting_room` | `bot.in_waiting_room` | `InWaitingRoom` | 200 |
189
+ | `bot.in_call_not_recording` | `bot.inmeeting` | `InMeeting` | 200 |
190
+ | `bot.in_call_recording` | `bot.recording` | `Recording` | 200 |
191
+ | _(none — new)_ | `bot.leaving` | `Leaving` | 200 |
192
+ | _(none — new)_ | `bot.error` (non-terminal streaming-provider error; bot keeps running) | `InMeeting` | — (omitted) |
193
+ | `bot.call_ended` | `bot.stopped` | `Stopped` | 200 |
194
+ | `bot.call_ended` (lobby timeout) | `bot.stopped` | `NotAllowed` | 500 |
195
+ | `bot.call_ended` (host denied) | `bot.stopped` | `Denied` | 500 |
196
+ | `bot.fatal` | `bot.stopped` | `Error` | 500 |
197
+ | `bot.fatal` (`meeting_requires_sign_in`) | `bot.stopped` (`bot_event: bot.failed`) | `Failed` — `ErrorReason` `SignInRequired` (Google Meet, Teams) | branch on `bot_event`, not `status_code` |
198
+ | `bot.fatal` (`meeting_not_found`) | `bot.stopped` (`bot_event: bot.failed`) | `Failed` — `ErrorReason` `InvalidMeetingLink` (Google Meet, Teams) | branch on `bot_event`, not `status_code` |
199
+ | _(none — new)_ | `bot.stopped` (`bot_event: bot.failed`) | `Failed` — `ErrorReason` `MeetingRestricted` or `CantJoinCall` (Google Meet; webhook adds `failure_reason`) | branch on `bot_event`, not `status_code` |
200
+ | `bot.recording_permission_allowed` | `bot.recording_permission_allowed` (Zoom) | `RecordingPermissionAllowed` | — |
201
+ | `bot.recording_permission_denied` | `bot.recording_permission_denied` (Zoom; bot leaves after `recording_permission_denied_timeout`, 60-600 s, default 600) | `RecordingPermissionDenied` | 500 |
202
+ | `recording.done` | `manifest.completed` → `audio.processed` → `video.processed` (if `video_required`) | — | 200 |
203
+ | `transcript.done` | `transcription.processed` | — | 200 |
204
+ | `transcript.failed` | `transcription.failed` | — | **500** |
205
+ | _(post-call)_ | `bot.done` — post-call providers only | `Done` | 200 / 500 |
206
+ | _(retention)_ | `data_deletion` | — | 200 |
207
+
208
+ **Streaming-only providers** (`deepgram_streaming`, `assemblyai_streaming`, `meeting_captions`) do **not** fire `transcription.processed`/`transcription.failed`/`bot.done` — their terminal event is `audio.processed`. Don't wait on `bot.done` for a streaming bot, or your handler hangs forever.
209
+
210
+ **`transcript_id` is never in the webhook payload.** Fetch the transcript by bot with `GET /bots/{id}/get_transcript`, or resolve the ID via `GET /bots/{id}/detail` (`bot_details.transcript_id`), the `create_bot` response, or `GET /bots/{id}/transcriptions`.
211
+
212
+ **Common `bot_status` values:** `Joining`, `InWaitingRoom`, `InMeeting`, `Recording`, `Leaving`, `Stopped`, `NotAllowed`, `Denied`, `Failed`, `Error`, `Done`, and on Zoom `RecordingPermissionAllowed` / `RecordingPermissionDenied`. The list can grow: handle unknown values gracefully. `ErrorReason` (on `GET /bots/{id}/detail`) gives the exact cause.
213
+
214
+ ### Real-time events
215
+
216
+ | Recall.ai | MeetStream | Notes |
217
+ |-----------|------------|-------|
218
+ | `transcript.data` / `transcript.partial_data` | delivered to `live_transcription_required.webhook_url` | Chunk shape: `{ speakerName, timestamp, transcript, words[] }` |
219
+ | `participant_events.join` / `.leave` | same event names, requires `recording_config.realtime_endpoints` | Different payload shape — nested under `data.bot.{id,metadata}`, no top-level `bot_id` |
220
+ | _(none)_ | `participant_events.raise_hand` / `.lower_hand` | Opt-in: list them in `recording_config.realtime_endpoints[].events`; same payload shape as `.join` |
221
+
222
+ ---
223
+
224
+ ## Breaking changes to review manually
225
+
226
+ 1. **Transcript retrieval**: `GET /bots/{id}/get_transcript` takes the `bot_id`. Use `GET /transcript/{transcript_id}/get_transcript` for a specific transcript.
227
+ 2. **`recording_mode`** (Recall string) → **`video_required`** (MeetStream boolean).
228
+ 3. **Chat messages**: Recall's separate `POST /send_chat_message/` → MeetStream's `POST /bots/{id}/send_message`.
229
+ 4. **Webhook registration**: Recall registers globally via `POST /webhook/`; MeetStream uses `callback_url` per bot at creation.
230
+ 5. **Webhook event values** rename (see table above) — but the **envelope key stays `event`**, so most handler code needs a value rename, not a restructure.
231
+ 6. **Post-call processing** arrives as discrete events (`manifest.completed`, `audio.processed`, `transcription.processed`/`transcription.failed`, `video.processed`, `bot.done`) instead of one `bot.done`-then-fetch signal.
232
+ 7. **Live transcription**: `real_time_transcription.destination_url` → `live_transcription_required.webhook_url` on `create_bot`.
233
+ 8. **Pause/resume recording** maps 1:1 (`POST /bots/{id}/pause_recording|resume_recording`, empty body).
234
+ 9. **Output audio/media** into the call is REST on Recall. On MeetStream, audio goes over the WebSocket bridge and live camera video over `bot_camera_stream` (WebRTC) — needs a real refactor, not a rename (flagged, not auto-migrated).
235
+ 10. **Zoom** needs no setup: remove Zoom app, SDK key or join-token plumbing unless you want signed-in joins (`zoom.zak_url` / `zoom.obf_url` with your own Zoom app).
236
+
237
+ ## Post-Migration Checklist
238
+
239
+ 1. Set `MEETSTREAM_API_KEY` (get one at [app.meetstream.ai/api-keys](https://app.meetstream.ai/api-keys))
240
+ 2. Review every ⚠️ warning from the migration output
241
+ 3. Update webhook handlers: keep branching on `event`, rename the values, and branch on `bot_event` for `bot.stopped`'s reason (`bot.kicked`, `bot.notallowed`, `bot.denied`, `bot.failed`)
242
+ 4. Fetch transcripts by bot (`GET /bots/{id}/get_transcript`) or by `transcript_id`
243
+ 5. Add handlers for `manifest.completed`, `audio.processed`, `transcription.processed`/`failed`, `video.processed`
244
+ 6. Verify the `recording_mode` → `video_required` boolean logic
245
+ 7. Keep `pause_recording`/`resume_recording` calls (paths are rewritten 1:1)
246
+ 8. Send the key as `Authorization: Token <key>`: the `Token ` prefix is required on MeetStream (optional on Recall).
247
+ 9. Run `meetstream-migrate gaps` and plan the Recall features that have no MeetStream mapping.
248
+ 10. Run `meetstream-migrate test --api-key YOUR_KEY` to confirm connectivity
249
+
250
+ ---
251
+
252
+ ## Feature parity & gaps
253
+
254
+ | Recall capability | MeetStream | Migrator behaviour |
255
+ |---|---|---|
256
+ | Create/list/get/remove/delete bot | ✅ | auto-mapped |
257
+ | Recording (audio/video) + retention | ✅ | auto-mapped |
258
+ | Per-participant streams | ✅ | `get_audio_streams` / `get_recording_streams` |
259
+ | Real-time audio / transcript | ✅ | `live_audio_required` / `live_transcription_required` (live video is not delivered) |
260
+ | Send chat message | ✅ | `send_chat_message` → REST `send_message` |
261
+ | Output image/camera frame | ✅ | `output_video` → `send_image` (base64 `img` or public URL) |
262
+ | Calendar (Google + Outlook) + scheduling | ✅ | `calendar/events/{id}/bot` → `calendar/schedule/{id}` |
263
+ | Zoom bots | ✅ no setup | API key + meeting link; optional `zoom.zak_url` / `zoom.obf_url` with your own Zoom app (`zoom.use_own_app`) |
264
+ | Signed-in / OAuth bots | ✅ | Google login domains/logins; Zoom ZAK/OBF (`zoom.zak_url` / `zoom.obf_url`) |
265
+ | AI summaries | Build from the transcript | MeetStream has no summary endpoint: fetch the transcript and summarize it with your own model |
266
+ | Conversational voice agents (MIA) | ✅ (MeetStream-only) | `agent_config_id` on `create_bot` |
267
+ | Output audio / interactive media | ⚠️ manual | audio → WebSocket bridge; video → `bot_camera_stream` (flagged) |
268
+ | Pause/resume recording | ✅ 1:1 | `POST /bots/{id}/pause_recording` + `resume_recording` (empty body) |
269
+ | Live camera video into the call | ✅ (manual) | `output_media` camera → `bot_camera_stream.websocket_url` (WebRTC publisher on your side; Google Meet, Teams, Zoom) |
270
+ | Raise / lower the bot's hand | ✅ (MeetStream-only) | `POST /bots/{id}/send_command` with `{"command": "raise_hand"}` or `"lower_hand"` (Google Meet, Zoom, Teams) |
271
+ | Recordings to your own bucket | ✅ (MeetStream-only) | `storage_id` on `create_bot` selects a storage destination saved in the dashboard |
272
+
273
+ Run `scan` first to see exactly which of these appear in your codebase before migrating.
274
+
275
+ ---
276
+
277
+ ## Recall features with no MeetStream mapping
278
+
279
+ The migrator does not rewrite these. `scan` flags the ones it finds in your code, and `meetstream-migrate gaps` prints this list.
280
+
281
+ | Recall.ai feature | On MeetStream |
282
+ |---|---|
283
+ | Desktop Recording SDK (sdk_upload) | No equivalent. |
284
+ | Mobile Recording SDK | No equivalent. |
285
+ | Webex, GoTo Meeting, Slack huddle bots | MeetStream bots join Zoom, Google Meet and Microsoft Teams (WhatsApp calls in beta). Links from other services return 400. |
286
+ | Regions (us-west-2, eu-central-1, ap-northeast-1) | One base URL, https://api.meetstream.ai/api/v1. |
287
+ | output_audio mid-call (REST) | No REST call: socket_connection_url command sendaudio, or initial_audio once at join. |
288
+ | output_screenshare, pin_participant | No equivalent. On Google Meet a voice agent can present a live web page (agent.tools: ["screen"]). |
289
+ | output_media camera webpage, and output_media audio with third-party agents (LiveKit, Vapi, ElevenLabs, Pipecat, OpenAI) | Not 1:1. Use the Agents API (voice agent on the bot, your own provider keys or your own agent via local_agent), or your own bridge with live_audio_required and socket_connection_url, or bot_camera_stream for your own video. |
290
+ | start_recording / stop_recording, recording_config.start_recording_on | Records from admission; pause_recording / resume_recording. |
291
+ | include_bot_in_recording, video_mixed_participant_video_when_screenshare, variant | No equivalent. |
292
+ | Real-time: rtmp endpoints, video_separate_png / video_separate_h264, audio_separate_raw, participant speech_on/off, webcam, microphone and screenshare events | No live equivalent. After the call: get_speaker_timeline, get_audio_streams, get_recording_streams. |
293
+ | breakout_room | No equivalent. |
294
+ | automatic_leave.silence_detection, in_call_not_recording_timeout, everyone_left_timeout.activate_after | No equivalent. everyone_left_timeout takes a number of seconds. |
295
+ | automatic_audio_output.replay_on_participant_join | initial_audio plays once. |
296
+ | chat.on_participant_join, send_chat_message to / pin | bot_message once at join; send_message posts to everyone. |
297
+ | Calendar V1 (calendar users, should_record, refresh), calendar access-token, calendar.update / calendar.sync_events webhooks | No 1:1 match. Connect calendars and use auto-schedule or per-event scheduling. |
298
+ | Zoom OAuth apps / credentials API and logs | Dashboard > Integrations > Zoom. |
299
+ | Meeting Direct Connect for Google Meet Media API | Zoom (RTMS) only. |
300
+ | Billing usage endpoint | Dashboard. GET /api/v1/billing/balance returns the prepaid balance where it is on for the account. |
301
+ | Per-artifact PATCH (metadata) and per-artifact DELETE | custom_attributes are set on create_bot. DELETE /api/v1/bots/{bot_id}/delete removes all media of a bot. |
302
+ | Streaming transcription providers gladia_v2, rev, aws_transcribe, speechmatics, elevenlabs, soniox, grok | Not offered. MeetStream providers: deepgram, assemblyai, sarvam, meetstream, jigsawstack and smallest (each also as _streaming), and meeting_captions. |
303
+ | Webhook retries (Svix retry schedule) | One delivery for each event. Reply 2xx fast. |
304
+
305
+ ## Latest docs
306
+
307
+ - [Migrate from Recall.ai](https://docs.meetstream.ai/migration/migrate-from-recall)
308
+ - [Two ways to integrate](https://docs.meetstream.ai/guides/introduction/integration-paths) · [Create Bot payload reference](https://docs.meetstream.ai/api-reference/create-bot-payload-reference)
309
+ - [Webhooks and events](https://docs.meetstream.ai/guides/webhooks/webhooks-and-events) · [Webhook signature verification](https://docs.meetstream.ai/guides/webhooks/webhook-signature-verification) · [Errors](https://docs.meetstream.ai/errors)
310
+
311
+ For the latest details, fetch the page with `.md` appended, read https://docs.meetstream.ai/llms.txt, or connect the docs MCP server at https://docs.meetstream.ai/_mcp/server. If this README and the docs differ, the docs are correct.
312
+
313
+ ## Requirements
314
+
315
+ Python ≥ 3.8
316
+
317
+ ## Development
318
+
319
+ ```bash
320
+ pip install -e .
321
+ python -m pytest test/ # or: python -m meetstream_migrate.cli test --api-key YOUR_KEY
322
+ ```
323
+
324
+ ## License
325
+
326
+ MIT © MeetStream.ai — [meetstream.ai](https://meetstream.ai) · [docs.meetstream.ai](https://docs.meetstream.ai)
@@ -0,0 +1,290 @@
1
+ # meetstream-migrate
2
+
3
+ **One-command migration from [Recall.ai](https://recall.ai) to [MeetStream.ai](https://meetstream.ai).** Scans your codebase, maps every API call, shows the diff, and migrates — safely, with warnings for anything that needs a human look.
4
+
5
+ > **v2.5.0 — mappings for the MeetStream API (v1.10.0) and the Recall.ai API v1.11, November 2026.**
6
+ > MeetStream matches Recall across recording, per-participant streams, real-time audio/video/
7
+ > transcript, chat + image output, Google **and** Outlook calendar, scheduling, signed-in bots,
8
+ > and retention — plus voice agents in the meeting (Agents API) that Recall doesn't have.
9
+ > Pause/resume recording maps 1:1, Zoom bots need no setup (API key + meeting link), and live
10
+ > camera video into the call uses `bot_camera_stream` (WebRTC). See
11
+ > [Feature parity & gaps](#feature-parity--gaps) below.
12
+
13
+ ---
14
+
15
+ **Two ways to integrate MeetStream.** (1) The **Bot API**: meeting bots and recording, the direct replacement for Recall bots. (2) The **Agents API**: a voice agent (API name: MIA) on the bot, which talks in the meeting. This migrator moves Recall code to the Bot API; add an agent later with `agent_config_id` on `create_bot`. Guide: [Two ways to integrate](https://docs.meetstream.ai/guides/introduction/integration-paths).
16
+
17
+ ## Quick Start
18
+
19
+ ```bash
20
+ pip install meetstream-migrate
21
+ meetstream-migrate scan ./your-recall-project
22
+ ```
23
+
24
+ ## Commands
25
+
26
+ ```bash
27
+ # Scan (read-only) — finds every Recall.ai reference, writes NOTHING
28
+ meetstream-migrate scan ./your-project
29
+
30
+ # Dry run — shows exactly what would change, still writes nothing
31
+ meetstream-migrate migrate ./your-project --dry-run
32
+
33
+ # Migrate — applies all transformations (prompts before writing)
34
+ meetstream-migrate migrate ./your-project --yes
35
+
36
+ # List Recall.ai features with no MeetStream mapping
37
+ meetstream-migrate gaps
38
+
39
+ # Verify your MeetStream credentials + API reachability
40
+ meetstream-migrate test --api-key YOUR_MEETSTREAM_API_KEY
41
+ ```
42
+
43
+ ### CLI reference
44
+
45
+ ```
46
+ meetstream-migrate <command> [path] [options]
47
+
48
+ Commands:
49
+ scan [path] Scan only — show what would change
50
+ migrate [path] Scan & migrate a project
51
+ test Run API tests against MeetStream.ai
52
+ gaps List Recall.ai features with no MeetStream mapping
53
+
54
+ Options:
55
+ --yes, -y Skip the confirmation prompt
56
+ --dry-run Show changes without writing files
57
+ --api-key KEY MeetStream API key (for the test command)
58
+ --verbose, -v Detailed output
59
+ --help, -h Show help
60
+ ```
61
+
62
+ ## What it does
63
+
64
+ **1. Scan** — recursively finds:
65
+ - Recall.ai base URLs (`us-east-1.recall.ai`, `eu-west-1.recall.ai`, `api.recall.ai`)
66
+ - API endpoint calls (`/api/v1/bot/`, `/api/v1/bot/{id}/audio/`, etc.)
67
+ - Recall-specific field names (`meeting_url`, `recording_mode`, `transcription_options`, `metadata`, …)
68
+ - Webhook handlers (`bot.status_change`, `joining_call`, `in_call_recording`, `call_ended`, `done`, `fatal`, …)
69
+ - SDK imports (`require('recall-ai')`, `import recall`)
70
+ - Environment variables (`RECALL_API_KEY`, `RECALL_API_URL`, `RECALL_WEBHOOK_SECRET`)
71
+
72
+ Supported file types: `.js .ts .jsx .tsx .mjs .cjs .py .go .rb .java .kt .rs .php .cs .swift .env .yaml .yml .json .toml .ini .cfg`
73
+
74
+ **2. Display** — endpoint mapping table, field mapping table, webhook event mapping, and a compatibility check (auto-migratable vs. needs manual review).
75
+
76
+ **3. Transform** — base URL swap, endpoint path rewrites, field renames, webhook event renames, env var renames, import/require updates. Every change carries an inline warning when it needs a second look.
77
+
78
+ **4. Report** — what changed, what needs manual review, and a post-migration checklist.
79
+
80
+ ---
81
+
82
+ ## Endpoint Mapping
83
+
84
+ | Recall.ai | MeetStream.ai | Notes |
85
+ |-----------|---------------|-------|
86
+ | `POST /api/v1/bot/` | `POST /api/v1/bots/create_bot` | Field names differ (see below) |
87
+ | `GET /api/v1/bot/{id}/` | `GET /api/v1/bots/{id}/detail` | Also `GET /api/v1/bots/{id}/status` |
88
+ | `POST /api/v1/bot/{id}/leave_call/` | `GET /api/v1/bots/{id}/remove_bot` | ⚠️ Method changes POST → GET |
89
+ | `GET /api/v1/bot/` | `GET /api/v1/bots` | List all bots |
90
+ | `GET /bot/{id}/audio/` | `GET /bots/{id}/get_audio` | Download link, valid 1 hour |
91
+ | `GET /bot/{id}/video/` | `GET /bots/{id}/get_video` | Download link, valid 1 hour |
92
+ | `GET /bot/{id}/transcript/` | `GET /bots/{id}/get_transcript` | Takes the `bot_id`; for a specific transcript use `GET /transcript/{transcript_id}/get_transcript` |
93
+ | `GET /bot/{id}/participants/` | `GET /bots/{id}/get_participants` | |
94
+ | `GET /bot/{id}/chat_messages/` | `GET /bots/{id}/get_chats` | |
95
+ | `POST /bot/{id}/send_chat_message/` | `POST /bots/{id}/send_message` | REST body `{ "message": "..." }` — drop Recall's `to`/`pin` fields |
96
+ | `POST /bot/{id}/output_video/` | `POST /bots/{id}/send_image` | `img` (base64, from `b64_data`) or a public `img_url` |
97
+ | `POST /bot/{id}/output_audio/` | WebSocket `sendaudio` command | ⚠️ Manual — no REST equivalent, use `socket_connection_url` bridge |
98
+ | `POST /bot/{id}/output_media/` | `bot_camera_stream.websocket_url` on `create_bot` | ⚠️ Manual — publish WebRTC video through a `wss://` signaling socket; no webpage field |
99
+ | `POST /bot/{id}/pause_recording/` | `POST /bots/{id}/pause_recording` | ✅ 1:1 (empty body) |
100
+ | `POST /bot/{id}/resume_recording/` | `POST /bots/{id}/resume_recording` | ✅ 1:1 (empty body) |
101
+ | `GET /bot/{id}/screenshots/` | `GET /bots/{id}/screenshots` | |
102
+ | `GET /bot/{id}/speaker_timeline/` | `GET /bots/{id}/get_speaker_timeline` | Large timelines return `speaker_timeline_url` (download link, valid 1 h) instead of inline JSON |
103
+ | `POST /bot/{id}/delete_media/` | `DELETE /bots/{id}/delete` | ⚠️ Method POST → DELETE; permanent |
104
+ | `PATCH /api/v1/bot/{id}/` (scheduled bot) | `PATCH /api/v1/calendar/scheduled_bots/{bot_id}` | New `scheduled_join_time` and other create fields; only while `Scheduled` |
105
+ | `DELETE /api/v1/bot/{id}/` (scheduled bot) | `DELETE /api/v1/calendar/scheduled_bots/{bot_id}` | To delete media instead: `DELETE /bots/{id}/delete` |
106
+ | `POST /recording/{id}/create_transcript/` | `POST /bots/{bot_id}/transcribe` | Keyed by `bot_id`; list runs with `GET /bots/{bot_id}/transcriptions` (flagged) |
107
+ | `GET /recording/{id}/`, `/meeting_metadata/{id}/` | `GET /bots/{bot_id}/detail` | One bot = one recording (flagged) |
108
+ | `GET /transcript/{id}/` | `GET /transcript/{transcript_id}/get_transcript` | Returns segments; a failed transcript is `422` with the reason (flagged) |
109
+ | `GET /audio_mixed/`, `/video_mixed/` | `GET /bots/{bot_id}/get_audio`, `get_video` | flagged |
110
+ | `GET /audio_separate/`, `/video_separate/` | `GET /bots/{bot_id}/get_audio_streams`, `get_recording_streams` | Needs `audio_separate_streams` / `video_separate_streams` (flagged) |
111
+ | `GET /participant_events/` | `GET /bots/{bot_id}/get_participants`, `get_speaker_timeline` | flagged |
112
+ | `POST`/`DELETE /api/v2/calendar-events/{id}/bot/` | `POST`/`DELETE /api/v1/calendar/schedule/{event_id}` | `bot_config` fields go in the body; `deduplication_key` works |
113
+ | `GET /api/v2/calendar-events/` | `GET /api/v1/calendar/events` | |
114
+ | `/api/v2/calendars/` | `GET /api/v1/calendar/calendars`; create `POST /api/v1/calendar/create_calendar` or `create_outlook_calendar`; `DELETE /api/v1/calendar/disconnect` | ⚠️ Create and delete need manual edits |
115
+ | `POST /api/v1/meeting_direct_connect/` | `POST /api/v1/meeting_direct_connect` | Zoom (RTMS) only |
116
+ | `/api/v2/google-login-groups/`, `/api/v2/google-logins/` | `/api/v1/google-login-domains`, `/api/v1/google-logins` | ⚠️ Bodies differ (SSO key and certificate) |
117
+ | `start_recording`, `stop_recording`, `output_screenshare`, `pin_participant`, `billing/usage`, `zoom-oauth-*`, `sdk_upload` | none | Flagged by `scan`; see `gaps` |
118
+ | `POST /api/v1/webhook/` | Workspace webhooks (Dashboard > Configure > Webhooks, signed) or `callback_url` on `create_bot` (not signed) | No API call registers webhooks |
119
+ | `GET /bot/{id}/calendar/…` | `POST /calendar/create_calendar` (underscore) | Body: `google_client_id`/`google_client_secret`/`google_refresh_token` |
120
+
121
+ ## Field Mapping
122
+
123
+ | Recall.ai | MeetStream.ai | Notes |
124
+ |-----------|---------------|-------|
125
+ | `meeting_url` | `meeting_link` | |
126
+ | `recording_mode` (string) | `video_required` (boolean) | `"speaker_view"`/`"gallery_view"` → `true`, `"audio_only"` → `false` |
127
+ | `metadata` | `custom_attributes` | Values must be strings |
128
+ | `transcription_options` | `recording_config.transcript.provider.{provider}` | Object structure differs — see provider config |
129
+ | `noone_joined_timeout` | `no_one_joined_timeout` | |
130
+ | `recording_permission_denied_timeout` | same name in `automatic_leave` | Zoom only; 60-600 s, default 600 |
131
+ | `zoom.zak_url` / `zoom.obf_token_url` | `zoom.zak_url` / `zoom.obf_url` | Optional; uses your own Zoom app (`use_own_app: true`) |
132
+ | `zoom.join_token_url` | — | Remove: Zoom bots join with no setup |
133
+ | `destination_url` | `webhook_url` (nested in `live_transcription_required`) | |
134
+ | `assembly_ai` | `assemblyai` | Also: `assemblyai` uses `speech_models[]` + `language_code`, not `model`/`language` |
135
+ | `output_media.websocket.url` | `socket_connection_url.websocket_url` | |
136
+ | `output_media.camera` (webpage) | `bot_camera_stream.websocket_url` | ⚠️ Manual — you publish WebRTC video; MeetStream does not render a webpage |
137
+
138
+ ## Webhook Events — live-verified against production
139
+
140
+ MeetStream's envelope key is **`event`** — the **same key Recall uses**. No envelope rename needed. Alongside it: `bot_id`, `message`, `status_code` (200 success / 500 failure), `custom_attributes`, and on lifecycle events, `bot_status`.
141
+
142
+ ```json
143
+ { "event": "bot.inmeeting", "bot_id": "...", "bot_status": "InMeeting",
144
+ "message": "...", "status_code": 200, "custom_attributes": {} }
145
+ ```
146
+
147
+ **Two-layer model:** `bot.stopped` fires **once**; `bot_status` tells you *why* — `Stopped` (normal) / `NotAllowed` (lobby timeout) / `Denied` (host denied) / `Failed` (the meeting blocked the join) / `Error` (crash). `bot_event` gives the detail (`bot.kicked`, `bot.denied`, `bot.notallowed`, `bot.failed`). Recall's separate end-reason events collapse into this single event + status.
148
+
149
+ | Recall.ai Event | MeetStream `event` | `bot_status` | status_code |
150
+ |----------------|--------------------|--------------|-------------|
151
+ | `bot.joining_call` | `bot.joining` (may fire up to 3×) | `Joining` | 200 |
152
+ | `bot.in_waiting_room` | `bot.in_waiting_room` | `InWaitingRoom` | 200 |
153
+ | `bot.in_call_not_recording` | `bot.inmeeting` | `InMeeting` | 200 |
154
+ | `bot.in_call_recording` | `bot.recording` | `Recording` | 200 |
155
+ | _(none — new)_ | `bot.leaving` | `Leaving` | 200 |
156
+ | _(none — new)_ | `bot.error` (non-terminal streaming-provider error; bot keeps running) | `InMeeting` | — (omitted) |
157
+ | `bot.call_ended` | `bot.stopped` | `Stopped` | 200 |
158
+ | `bot.call_ended` (lobby timeout) | `bot.stopped` | `NotAllowed` | 500 |
159
+ | `bot.call_ended` (host denied) | `bot.stopped` | `Denied` | 500 |
160
+ | `bot.fatal` | `bot.stopped` | `Error` | 500 |
161
+ | `bot.fatal` (`meeting_requires_sign_in`) | `bot.stopped` (`bot_event: bot.failed`) | `Failed` — `ErrorReason` `SignInRequired` (Google Meet, Teams) | branch on `bot_event`, not `status_code` |
162
+ | `bot.fatal` (`meeting_not_found`) | `bot.stopped` (`bot_event: bot.failed`) | `Failed` — `ErrorReason` `InvalidMeetingLink` (Google Meet, Teams) | branch on `bot_event`, not `status_code` |
163
+ | _(none — new)_ | `bot.stopped` (`bot_event: bot.failed`) | `Failed` — `ErrorReason` `MeetingRestricted` or `CantJoinCall` (Google Meet; webhook adds `failure_reason`) | branch on `bot_event`, not `status_code` |
164
+ | `bot.recording_permission_allowed` | `bot.recording_permission_allowed` (Zoom) | `RecordingPermissionAllowed` | — |
165
+ | `bot.recording_permission_denied` | `bot.recording_permission_denied` (Zoom; bot leaves after `recording_permission_denied_timeout`, 60-600 s, default 600) | `RecordingPermissionDenied` | 500 |
166
+ | `recording.done` | `manifest.completed` → `audio.processed` → `video.processed` (if `video_required`) | — | 200 |
167
+ | `transcript.done` | `transcription.processed` | — | 200 |
168
+ | `transcript.failed` | `transcription.failed` | — | **500** |
169
+ | _(post-call)_ | `bot.done` — post-call providers only | `Done` | 200 / 500 |
170
+ | _(retention)_ | `data_deletion` | — | 200 |
171
+
172
+ **Streaming-only providers** (`deepgram_streaming`, `assemblyai_streaming`, `meeting_captions`) do **not** fire `transcription.processed`/`transcription.failed`/`bot.done` — their terminal event is `audio.processed`. Don't wait on `bot.done` for a streaming bot, or your handler hangs forever.
173
+
174
+ **`transcript_id` is never in the webhook payload.** Fetch the transcript by bot with `GET /bots/{id}/get_transcript`, or resolve the ID via `GET /bots/{id}/detail` (`bot_details.transcript_id`), the `create_bot` response, or `GET /bots/{id}/transcriptions`.
175
+
176
+ **Common `bot_status` values:** `Joining`, `InWaitingRoom`, `InMeeting`, `Recording`, `Leaving`, `Stopped`, `NotAllowed`, `Denied`, `Failed`, `Error`, `Done`, and on Zoom `RecordingPermissionAllowed` / `RecordingPermissionDenied`. The list can grow: handle unknown values gracefully. `ErrorReason` (on `GET /bots/{id}/detail`) gives the exact cause.
177
+
178
+ ### Real-time events
179
+
180
+ | Recall.ai | MeetStream | Notes |
181
+ |-----------|------------|-------|
182
+ | `transcript.data` / `transcript.partial_data` | delivered to `live_transcription_required.webhook_url` | Chunk shape: `{ speakerName, timestamp, transcript, words[] }` |
183
+ | `participant_events.join` / `.leave` | same event names, requires `recording_config.realtime_endpoints` | Different payload shape — nested under `data.bot.{id,metadata}`, no top-level `bot_id` |
184
+ | _(none)_ | `participant_events.raise_hand` / `.lower_hand` | Opt-in: list them in `recording_config.realtime_endpoints[].events`; same payload shape as `.join` |
185
+
186
+ ---
187
+
188
+ ## Breaking changes to review manually
189
+
190
+ 1. **Transcript retrieval**: `GET /bots/{id}/get_transcript` takes the `bot_id`. Use `GET /transcript/{transcript_id}/get_transcript` for a specific transcript.
191
+ 2. **`recording_mode`** (Recall string) → **`video_required`** (MeetStream boolean).
192
+ 3. **Chat messages**: Recall's separate `POST /send_chat_message/` → MeetStream's `POST /bots/{id}/send_message`.
193
+ 4. **Webhook registration**: Recall registers globally via `POST /webhook/`; MeetStream uses `callback_url` per bot at creation.
194
+ 5. **Webhook event values** rename (see table above) — but the **envelope key stays `event`**, so most handler code needs a value rename, not a restructure.
195
+ 6. **Post-call processing** arrives as discrete events (`manifest.completed`, `audio.processed`, `transcription.processed`/`transcription.failed`, `video.processed`, `bot.done`) instead of one `bot.done`-then-fetch signal.
196
+ 7. **Live transcription**: `real_time_transcription.destination_url` → `live_transcription_required.webhook_url` on `create_bot`.
197
+ 8. **Pause/resume recording** maps 1:1 (`POST /bots/{id}/pause_recording|resume_recording`, empty body).
198
+ 9. **Output audio/media** into the call is REST on Recall. On MeetStream, audio goes over the WebSocket bridge and live camera video over `bot_camera_stream` (WebRTC) — needs a real refactor, not a rename (flagged, not auto-migrated).
199
+ 10. **Zoom** needs no setup: remove Zoom app, SDK key or join-token plumbing unless you want signed-in joins (`zoom.zak_url` / `zoom.obf_url` with your own Zoom app).
200
+
201
+ ## Post-Migration Checklist
202
+
203
+ 1. Set `MEETSTREAM_API_KEY` (get one at [app.meetstream.ai/api-keys](https://app.meetstream.ai/api-keys))
204
+ 2. Review every ⚠️ warning from the migration output
205
+ 3. Update webhook handlers: keep branching on `event`, rename the values, and branch on `bot_event` for `bot.stopped`'s reason (`bot.kicked`, `bot.notallowed`, `bot.denied`, `bot.failed`)
206
+ 4. Fetch transcripts by bot (`GET /bots/{id}/get_transcript`) or by `transcript_id`
207
+ 5. Add handlers for `manifest.completed`, `audio.processed`, `transcription.processed`/`failed`, `video.processed`
208
+ 6. Verify the `recording_mode` → `video_required` boolean logic
209
+ 7. Keep `pause_recording`/`resume_recording` calls (paths are rewritten 1:1)
210
+ 8. Send the key as `Authorization: Token <key>`: the `Token ` prefix is required on MeetStream (optional on Recall).
211
+ 9. Run `meetstream-migrate gaps` and plan the Recall features that have no MeetStream mapping.
212
+ 10. Run `meetstream-migrate test --api-key YOUR_KEY` to confirm connectivity
213
+
214
+ ---
215
+
216
+ ## Feature parity & gaps
217
+
218
+ | Recall capability | MeetStream | Migrator behaviour |
219
+ |---|---|---|
220
+ | Create/list/get/remove/delete bot | ✅ | auto-mapped |
221
+ | Recording (audio/video) + retention | ✅ | auto-mapped |
222
+ | Per-participant streams | ✅ | `get_audio_streams` / `get_recording_streams` |
223
+ | Real-time audio / transcript | ✅ | `live_audio_required` / `live_transcription_required` (live video is not delivered) |
224
+ | Send chat message | ✅ | `send_chat_message` → REST `send_message` |
225
+ | Output image/camera frame | ✅ | `output_video` → `send_image` (base64 `img` or public URL) |
226
+ | Calendar (Google + Outlook) + scheduling | ✅ | `calendar/events/{id}/bot` → `calendar/schedule/{id}` |
227
+ | Zoom bots | ✅ no setup | API key + meeting link; optional `zoom.zak_url` / `zoom.obf_url` with your own Zoom app (`zoom.use_own_app`) |
228
+ | Signed-in / OAuth bots | ✅ | Google login domains/logins; Zoom ZAK/OBF (`zoom.zak_url` / `zoom.obf_url`) |
229
+ | AI summaries | Build from the transcript | MeetStream has no summary endpoint: fetch the transcript and summarize it with your own model |
230
+ | Conversational voice agents (MIA) | ✅ (MeetStream-only) | `agent_config_id` on `create_bot` |
231
+ | Output audio / interactive media | ⚠️ manual | audio → WebSocket bridge; video → `bot_camera_stream` (flagged) |
232
+ | Pause/resume recording | ✅ 1:1 | `POST /bots/{id}/pause_recording` + `resume_recording` (empty body) |
233
+ | Live camera video into the call | ✅ (manual) | `output_media` camera → `bot_camera_stream.websocket_url` (WebRTC publisher on your side; Google Meet, Teams, Zoom) |
234
+ | Raise / lower the bot's hand | ✅ (MeetStream-only) | `POST /bots/{id}/send_command` with `{"command": "raise_hand"}` or `"lower_hand"` (Google Meet, Zoom, Teams) |
235
+ | Recordings to your own bucket | ✅ (MeetStream-only) | `storage_id` on `create_bot` selects a storage destination saved in the dashboard |
236
+
237
+ Run `scan` first to see exactly which of these appear in your codebase before migrating.
238
+
239
+ ---
240
+
241
+ ## Recall features with no MeetStream mapping
242
+
243
+ The migrator does not rewrite these. `scan` flags the ones it finds in your code, and `meetstream-migrate gaps` prints this list.
244
+
245
+ | Recall.ai feature | On MeetStream |
246
+ |---|---|
247
+ | Desktop Recording SDK (sdk_upload) | No equivalent. |
248
+ | Mobile Recording SDK | No equivalent. |
249
+ | Webex, GoTo Meeting, Slack huddle bots | MeetStream bots join Zoom, Google Meet and Microsoft Teams (WhatsApp calls in beta). Links from other services return 400. |
250
+ | Regions (us-west-2, eu-central-1, ap-northeast-1) | One base URL, https://api.meetstream.ai/api/v1. |
251
+ | output_audio mid-call (REST) | No REST call: socket_connection_url command sendaudio, or initial_audio once at join. |
252
+ | output_screenshare, pin_participant | No equivalent. On Google Meet a voice agent can present a live web page (agent.tools: ["screen"]). |
253
+ | output_media camera webpage, and output_media audio with third-party agents (LiveKit, Vapi, ElevenLabs, Pipecat, OpenAI) | Not 1:1. Use the Agents API (voice agent on the bot, your own provider keys or your own agent via local_agent), or your own bridge with live_audio_required and socket_connection_url, or bot_camera_stream for your own video. |
254
+ | start_recording / stop_recording, recording_config.start_recording_on | Records from admission; pause_recording / resume_recording. |
255
+ | include_bot_in_recording, video_mixed_participant_video_when_screenshare, variant | No equivalent. |
256
+ | Real-time: rtmp endpoints, video_separate_png / video_separate_h264, audio_separate_raw, participant speech_on/off, webcam, microphone and screenshare events | No live equivalent. After the call: get_speaker_timeline, get_audio_streams, get_recording_streams. |
257
+ | breakout_room | No equivalent. |
258
+ | automatic_leave.silence_detection, in_call_not_recording_timeout, everyone_left_timeout.activate_after | No equivalent. everyone_left_timeout takes a number of seconds. |
259
+ | automatic_audio_output.replay_on_participant_join | initial_audio plays once. |
260
+ | chat.on_participant_join, send_chat_message to / pin | bot_message once at join; send_message posts to everyone. |
261
+ | Calendar V1 (calendar users, should_record, refresh), calendar access-token, calendar.update / calendar.sync_events webhooks | No 1:1 match. Connect calendars and use auto-schedule or per-event scheduling. |
262
+ | Zoom OAuth apps / credentials API and logs | Dashboard > Integrations > Zoom. |
263
+ | Meeting Direct Connect for Google Meet Media API | Zoom (RTMS) only. |
264
+ | Billing usage endpoint | Dashboard. GET /api/v1/billing/balance returns the prepaid balance where it is on for the account. |
265
+ | Per-artifact PATCH (metadata) and per-artifact DELETE | custom_attributes are set on create_bot. DELETE /api/v1/bots/{bot_id}/delete removes all media of a bot. |
266
+ | Streaming transcription providers gladia_v2, rev, aws_transcribe, speechmatics, elevenlabs, soniox, grok | Not offered. MeetStream providers: deepgram, assemblyai, sarvam, meetstream, jigsawstack and smallest (each also as _streaming), and meeting_captions. |
267
+ | Webhook retries (Svix retry schedule) | One delivery for each event. Reply 2xx fast. |
268
+
269
+ ## Latest docs
270
+
271
+ - [Migrate from Recall.ai](https://docs.meetstream.ai/migration/migrate-from-recall)
272
+ - [Two ways to integrate](https://docs.meetstream.ai/guides/introduction/integration-paths) · [Create Bot payload reference](https://docs.meetstream.ai/api-reference/create-bot-payload-reference)
273
+ - [Webhooks and events](https://docs.meetstream.ai/guides/webhooks/webhooks-and-events) · [Webhook signature verification](https://docs.meetstream.ai/guides/webhooks/webhook-signature-verification) · [Errors](https://docs.meetstream.ai/errors)
274
+
275
+ For the latest details, fetch the page with `.md` appended, read https://docs.meetstream.ai/llms.txt, or connect the docs MCP server at https://docs.meetstream.ai/_mcp/server. If this README and the docs differ, the docs are correct.
276
+
277
+ ## Requirements
278
+
279
+ Python ≥ 3.8
280
+
281
+ ## Development
282
+
283
+ ```bash
284
+ pip install -e .
285
+ python -m pytest test/ # or: python -m meetstream_migrate.cli test --api-key YOUR_KEY
286
+ ```
287
+
288
+ ## License
289
+
290
+ MIT © MeetStream.ai — [meetstream.ai](https://meetstream.ai) · [docs.meetstream.ai](https://docs.meetstream.ai)
@@ -1,2 +1,2 @@
1
1
  """MeetStream Migration SDK — Recall.ai → MeetStream.ai"""
2
- __version__ = "2.3.0"
2
+ __version__ = "2.5.0"