meetstream-migrate 1.0.0

One-command migration from Recall.ai to MeetStream.ai — scans your codebase, maps APIs, transforms code.

meetstream-migrate

One-command migration from Recall.ai to MeetStream.ai. Scans your codebase, maps every API call, shows the diff, and migrates.

Why migrate?

• ~40% cheaper — significantly lower per-hour bot and transcription costs
• Agent-first architecture — built for AI agents joining meetings, not just recording
• Context layers — real-time info fetching during calls
• Migration CLI — scan, preview, migrate, test in minutes

Quick Start

pip install meetstream-migrate
meetstream-migrate scan ./your-recall-project

Or run without installing:

pipx run meetstream-migrate scan ./your-recall-project

Commands

Scan (read-only)

Finds every Recall.ai reference in your codebase. No files are modified.

meetstream-migrate scan ./your-project

Dry Run (preview changes)

Shows exactly what would change — still no writes.

meetstream-migrate migrate ./your-project --dry-run

Migrate

Applies all transformations. Prompts for confirmation before writing.

meetstream-migrate migrate ./your-project --yes

Test APIs

Verifies your MeetStream API key works and endpoints are reachable.

meetstream-migrate test --api-key YOUR_MEETSTREAM_API_KEY

What It Detects

The scanner finds:

• Base URLs — us-east-1.recall.ai, api.recall.ai
• API endpoints — all 14 Recall endpoints mapped to MeetStream equivalents
• Field names — meeting_url, recording_mode, metadata, transcription_options, etc.
• Webhook handlers — joining_call, in_call_recording, done, fatal, etc.
• SDK imports — import recall, from recall import
• Environment variables — RECALL_API_KEY

Supported file types: .js, .ts, .jsx, .tsx, .py, .go, .rb, .java, .env, .yaml, .json

Endpoint Mapping

Recall.ai	MeetStream.ai	Notes
POST /api/v1/bot/	POST /api/v1/bots/create_bot	Field names differ
GET /api/v1/bot/{id}/	GET /api/v1/bots/{id}/detail
POST /api/v1/bot/{id}/leave_call/	GET /api/v1/bots/{id}/remove_bot	Method change
DELETE /api/v1/bot/{id}/	DELETE /api/v1/bots/{id}/delete
GET /api/v1/bot/	GET /api/v1/bots	List all
GET /bot/{id}/audio/	GET /bots/{id}/get_audio
GET /bot/{id}/video/	GET /bots/{id}/get_video
GET /bot/{id}/transcript/	GET /transcript/{transcript_id}/get_transcript	Uses transcript_id
GET /bot/{id}/participants/	GET /bots/{id}/get_participants
GET /bot/{id}/chat_messages/	GET /bots/{id}/get_chats
POST /bot/{id}/send_chat_message/	bot_message field on create_bot	Manual review
GET /bot/{id}/screenshots/	GET /bots/{id}/get_screenshots
GET /bot/{id}/speaker_timeline/	GET /bots/{id}/get_speaker_timeline
POST /api/v1/webhook/	callback_url on create_bot	Per-bot webhooks

Field Mapping

Recall.ai	MeetStream.ai
meeting_url	meeting_link
recording_mode (string)	video_required (boolean)
metadata	custom_attributes
transcription_options	recording_config.transcript
noone_joined_timeout	voice_inactivity_timeout
destination_url	webhook_url
output_media.websocket.url	socket_connection_url.websocket_url

Webhook Event Mapping

IMPORTANT: MeetStream uses a two-layer event model, not a flat event list like Recall. You must check event first, then bot_status for detail — especially for bot.stopped where bot_status tells you why the bot stopped (normal exit, denied, lobby timeout, or error).

The webhook payload format is:

{ "event": "bot.inmeeting", "bot_id": "...", "bot_status": "InMeeting", "message": "...", "status_code": 200, "timestamp": "...", "custom_attributes": {} }

Bot Lifecycle Events

Recall.ai Event	MeetStream event	bot_status	Description
bot.joining_call	bot.joining	Joining	Bot connecting to meeting
bot.in_waiting_room	bot.joining	Joining	Bot in waiting room
bot.in_call_not_recording	bot.inmeeting	InMeeting	Joined, not yet recording
bot.recording_permission_allowed	bot.inmeeting	InMeeting	Recording allowed
bot.recording_permission_denied	bot.stopped	Denied	Recording denied by host
bot.in_call_recording	bot.inmeeting	InMeeting	In-call, recording
bot.call_ended	bot.stopped	Stopped	Meeting ended / bot left
bot.call_ended (lobby_timeout)	bot.stopped	NotAllowed	Bot could not join (lobby timeout)
bot.done	bot.stopped	Stopped	Bot lifecycle ended
bot.fatal	bot.stopped	Error	Unexpected error

Post-Call Processing Events (MeetStream)

Recall.ai	MeetStream event	Description
bot.done (then fetch)	audio.processed	Audio ready to fetch
bot.done (then fetch)	transcription.processed	Transcript ready to fetch
bot.done (then fetch)	video.processed	Video ready to fetch
(no equivalent)	data_deletion	Bot data deleted

Real-Time Events

Recall.ai Event	MeetStream	Description
transcript.data	live_transcription_required webhook	Transcript utterance
transcript.partial_data	live_transcription_required webhook	Partial transcript
transcript.done	transcription.processed	Full transcript ready
participant_events.join	Realtime endpoint	Participant joined
participant_events.leave	Realtime endpoint	Participant left

bot_status Reference

bot_status	event	Meaning
Joining	bot.joining	Bot is connecting
InMeeting	bot.inmeeting	Bot is in the meeting and recording
Stopped	bot.stopped	Bot exited normally (meeting ended, removed via API, everyone left, voice timeout, etc.)
NotAllowed	bot.stopped	Bot could not join (commonly waiting room / lobby timeout)
Denied	bot.stopped	Host explicitly denied join or recording permission
Error	bot.stopped	Unexpected error during lifecycle

Breaking Changes to Watch

These are flagged during migration and require manual review:

1. Transcript retrieval — MeetStream uses transcript_id (returned in create_bot response), not bot_id
2. Recording mode — Recall uses strings (speaker_view, gallery_view, audio_only), MeetStream uses a boolean video_required
3. Chat messages — Recall has a separate POST /send_chat_message/ endpoint, MeetStream uses the bot_message field on create_bot
4. Webhooks — Recall registers webhooks globally via POST /webhook/, MeetStream uses callback_url per bot on creation
5. Webhook events — Recall uses granular string events (bot.joining_call, bot.in_call_recording), MeetStream uses 3 lifecycle events (bot.joining, bot.inmeeting, bot.stopped) with bot_status for detail
6. Post-call processing — Recall signals readiness via bot.done, MeetStream sends separate audio.processed, transcription.processed, and video.processed webhooks
7. Live transcription — Recall uses real_time_transcription.destination_url, MeetStream uses live_transcription_required.webhook_url on create_bot

Post-Migration Checklist

After running the migration:

1. Set MEETSTREAM_API_KEY in your .env file (get one at app.meetstream.ai)
2. Review all warnings from the migration output
3. Update webhook handlers to use event + bot_status instead of Recall event names
4. Store transcript_id from create_bot responses for transcript retrieval
5. Add handlers for audio.processed, transcription.processed, video.processed webhooks
6. Verify recording_mode → video_required boolean logic
7. Run meetstream-migrate test --api-key YOUR_KEY to verify

CLI Options

meetstream-migrate [command] [path] [options]

Commands:
  scan [path]        Scan only — show what would change
  migrate [path]     Scan & migrate a project (default)
  test               Run API tests against MeetStream.ai

Options:
  --yes, -y          Skip confirmation prompt
  --dry-run          Show changes without writing files
  --api-key KEY      MeetStream API key (for test command)

Also available as an npm package

npx @meetstream/migrate scan ./your-project

See @meetstream/migrate on npm.

Requirements

• Python >= 3.8

License

MIT

---

Need help migrating? Reach out at meetstream.ai — we'll walk you through it.