## Background
This is team Comfortably Summed's submission for Twenty's Hacktoberfest.
We built an activity summary application that can periodically send
messages to the following platforms: Slack; Discord; and WhatsApp.
### Features
- 🧑💻 **People & Company Tracking**: Summarizes newly created people and
companies
- 🎯 **Opportunity Monitoring**: Reports on new opportunities created,
broken down by stage
- ✅ **Task Analytics**:
- Tracks task creation
- Calculates on-time completion rates
- Identifies team members with the most overdue tasks (the "slackers")
- 🔔 **Multi-Platform Notifications**: Send reports to Slack, Discord,
and/or WhatsApp
- ⏰ **Configurable Time Range**: Look back any number of days
### Summary of Changes
- Adds a new Twenty app called Activity Summary
- Contains a single index.ts file which utilises exported functions from
opportunity-creation-summariser.ts, people-creation-summariser.ts,
task-creation-summariser.ts, and senders.ts
- Implementation of sending a message to Slack, Discord, and WhatsApp
can be found in senders.ts
- Retrieval and summarising of Opportunity creation can be found in
opportunity-creation-summariser.ts
- Retrieval and summarising of People creation can be found in
people-creation-summariser.ts
- Retrieval and summarising of Task creation can be found in
task-creation-summariser.ts
## Screenshots
### Message to our Slack channel
<details>
<summary>Screenshot</summary>
<img width="326" height="242" alt="Screenshot 2025-11-01 at 22 05 30"
src="https://github.com/user-attachments/assets/57c5d50b-959d-4c3f-bd7d-00f42bf545b2"
/>
</details>
### Message to our Discord server's channel
<details>
<summary>Screenshot</summary>
<img width="472" height="386" alt="Screenshot 2025-11-01 at 22 06 44"
src="https://github.com/user-attachments/assets/f4a38d7f-e82d-47b0-a4b3-7bcf063fa575"
/>
</details>
### Message to our WhatsApp number
<details>
<summary>Screenshot</summary>
<img width="972" height="548" alt="IMG_2024"
src="https://github.com/user-attachments/assets/5533fc4d-a3ee-4e11-a9e7-9cc6a96316fc"
/>
</details>
### App-level configuration
<details>
<summary>Screenshot</summary>
<img width="442" height="385" alt="Screenshot 2025-11-01 at 22 02 14"
src="https://github.com/user-attachments/assets/c9948f57-f22c-42a0-972f-3348f480aa30"
/>
</details>
### Serverless functions
<details>
<summary>Screenshot</summary>
<img width="413" height="378" alt="Screenshot 2025-11-01 at 22 03 48"
src="https://github.com/user-attachments/assets/d297967b-52ce-4690-bb04-a16d89729d94"
/>
</details>
### Cron configuration
Default value in place due to Cron having a non-editable text input.
<details>
<summary>Screenshot</summary>
<img width="395" height="386" alt="Screenshot 2025-11-01 at 22 08 03"
src="https://github.com/user-attachments/assets/a95a708c-7136-4512-99c3-a6723adc0da5"
/>
</details>
## Testing
Sync the application to your Twenty instance and ensure the following
variables have values:
- `TWENTY_API_KEY` - Your Twenty CRM API key
- `DAYS_AGO` - Number of days to look back for the report
Choose any of the supported platforms and you shall see a summary being
sent!
# 🧠 AI-Powered Meeting Transcript to CRM Data Integration
## **Overview**
This feature automatically transforms meeting transcripts into
structured CRM data using AI.
When unstructured meeting notes are received via a **webhook**, the
system processes them and creates organized **notes, tasks, and
assignments** directly in **Twenty CRM**.
---
## **Key Features**
- **🤖 AI-Powered Analysis:**
Extracts **summaries, action items, assignees, and due dates** from
natural language transcripts.
- **📋 Smart Task Consolidation:**
Merges related sub-tasks into unified deliverables
*(e.g., `"draft" + "review" + "present"` → one consolidated task).*
- **👥 Intelligent Assignment:**
Uses **GraphQL member lookup** to match extracted assignee names to
workspace member IDs with flexible string matching.
- **🔗 Automatic Linking:**
Links generated **notes and tasks** to relevant contacts using
`noteTargets` and `taskTargets`.
- **🗓️ Date Parsing:**
Converts **relative date expressions** (e.g., “next Monday”, “end of
week”) into **ISO-formatted dates** for accurate scheduling.
---
## **Technical Stack**
| Component | Description |
|------------|-------------|
| **AI Provider** | Groq (via OpenAI SDK) using the `GPT-OSS-20B` model
|
| **APIs** | Twenty CRM REST API + GraphQL (for member resolution) |
| **Runtime** | Webhook-triggered **serverless function** written in
**TypeScript** |
---
## **Example Input**
```json
{
"transcript": "During the Project Phoenix Kick-off on November 1st, 2025, we discussed securing the Series B funding. ACTION: Dylan Field is designated to finalize the investor deck layout and needs to present it next Monday, November 4th. Irfan Hussain will review the deck before the presentation by Monday morning. COMMITMENT: Dario Amodei confirmed he would personally review the security protocols for the AI model before the end of this week, by Friday November 7th. Iqra Khan will coordinate the security review process and ensure completion by the Friday deadline.",
"meetingTitle": "Project Phoenix Kick-off",
"meetingDate": "2025-11-01",
"participants": [
"Brian Chesky",
"Dario Amodei",
"Iqra Khan",
"Irfan Hussain",
"Dylan Field"
],
"token": "e6d9d54e51953fd5a451cca933c63e7f8783b001f0c45be95be9d09ee06c6cda",
"relatedPersonId": "6c4b0e98-b69e-42a4-ba0c-fd2eeafca642"
}
```
---
## **Example Output**
```json
{
"success": true,
"noteId": "9cc3b4fc-ae37-4b3e-a343-a4c69cf6b1e8",
"taskIds": [
"0f408062-0dcc-49f0-9866-1ea05392661d",
"2b3739bf-0653-4101-9419-6a44ea5135cd"
],
"summary": {
"noteCreated": true,
"tasksCreated": 2,
"actionItemsProcessed": 2,
"commitmentsProcessed": 0
},
"executionLogs": [
"✅ Validation passed",
"📝 RelatedPersonId: 6c4b0e98-b69e-42a4-ba0c-fd2eeafca642",
"🤖 Starting transcript analysis...",
"✅ Analysis complete: 2 action items, 0 commitments",
"📄 Creating note in Twenty CRM...",
"✅ Note created: 9cc3b4fc-ae37-4b3e-a343-a4c69cf6b1e8",
"📋 Creating tasks from action items...",
"✅ Action item tasks created: 2",
"📋 Creating tasks from commitments...",
"✅ Commitment tasks created: 0"
]
}
```
---
Variable Name | Description
-- | --
GROQ_API_KEY | API key for authenticating requests to the Groq AI
service.
TWENTY_API_KEY | Authentication token used to access the Twenty CRM API.
TWENTY_API_URL | Base URL for the Twenty CRM REST API.
WEBHOOK_SECRET | Secret key used to validate incoming webhook requests
for security.
NODE_ENV | Defines the runtime environment (development, production,
etc.).
LOG_LEVEL | Controls verbosity of logs (info, debug, error).
---
## **Attachments**
<img width="649" height="863" alt="swappy-20251103-035128"
src="https://github.com/user-attachments/assets/2f0390af-9538-4fe2-bba8-f38e558935ad"
/>
---
https://github.com/user-attachments/assets/88620035-67ed-4150-b0be-46131083e2c5
---
Co-authored-by: iqra77818 <iqra77818@gmail.com>
This PR introduces an end-to-end workflow to automatically process
meeting transcripts and create structured notes and tasks in Twenty CRM.
It leverages OpenAI to extract summaries, key points, action items, and
participant commitments from transcripts.
Key features include:
1. AI-powered transcript analysis: Uses OpenAI GPT‑4o-mini to extract a
concise summary, key discussion points, action items, and commitments.
2. Automated note creation: Generates a rich Markdown note in Twenty CRM
with summary and key points.
3. Task automation: Automatically creates tasks in Twenty CRM from
extracted action items and commitments, linking them to the meeting
note.
4. Custom field support: Supports optional metadata from transcript
payloads, which can be included in note/task content or mapped to custom
fields in Twenty CRM if supported.
5. Webhook-ready: Designed to process Granola-style (or similar) webhook
payloads, making it easy to integrate with any AI meeting tool.
Sumbission for Hacktoberfest
Team Name : One for All
## Summary
- add packages/twenty-apps/rollup-engine: a parameterised rollup engine
that ships a default Opportunity → Company
aggregation.
- declare runtime config in package.json (TWENTY_API_KEY, optional
TWENTY_API_BASE_URL, and ROLLUP_ENGINE_CONFIG) so
app configuration lives entirely in Settings → Apps → [App] →
Configuration.
- document the workflow in README.md: deploy via twenty app sync,
populate env vars in the UI, use the Test panel with
a ready JSON payload, and reference troubleshooting tips.
- adjust the function entry point to a named async export and add
fallback logic for blank base URLs, matching the
UI’s env behaviour.
- prune legacy templates and examples so
config/templates/opportunity-to-company.json is the single copy/paste
starting point.
## UI / UX impact
After syncing the app:
- the Configuration screen shows the three env keys with helpful
descriptions (API key, optional base URL, JSON
override),
- the built-in Test your function panel works immediately
- and the default JSON config is available from
config/templates/opportunity-to-company.json for users who need to
customise rollups.
## Testing
- Out-of-the-box deploy to the hosted workspace (Opportunity update) ✓
- “Test your function” with the default config ✓
- Override example: point debugOpportunityCount at a scratch field via
ROLLUP_ENGINE_CONFIG ✓
- Optional: local smoke test (yarn install && yarn smoke) still passes
## Context
Having rollout feature for new workspaces creates bad experience for new
users and doesn't bring as much value as existing ones anyway. Removing
migration v2 from default feature flag and the whole concept of default
feature flag
# Introduction
When creating a view with v2 flag activated in production result in race
condition due to request being slow and //.
That's why we're introducing a batch create on view field here
closing https://github.com/twentyhq/core-team-issues/issues/1836
## In v2
- batch create view field endpoint is available
- frontend will target the new endpoint
## In v1
- batch create view field endpoint is not available
- frontend will stick to old fake batch view field creation loop
## Polish
- use persist view field is quite verbose as contains two data model we
could aim to create an update many view fields in order to standardize
new pattern
@charlesBochet had a conversation with Felix and he said we don't need
to spend time upgrading `zapier-platform-core` and `zapier-platform-cli`
since `twenty-zapier` will be deprecated anyway, making those packages
irrelevant.
I have updated tmp to a safer version elsewhere using `yarn up tmp
--recursive`. Also added `yarn.lock` to both server and front ci.
The massive changes in `yarn.lock` were introduced by
`zapier-platform-cli` version 17x - not sure if they caused those
breaking changes, but if they did and we still want update
zapier-related packages, I will take that up in another PR.
## Tests
### makeSureDashboardNamingAvailableCommand
Case 1: no dashboard custom object
Case 2: with dashboard custom object
### SeedDashboardViewCommand
Case 1: no existing view
Case 2: with existing view
changes -
- make iframe side panel to match others -- ie use SidePanelHeader for
title
- make url optional in configuration to match that of the other widgets
(allow partial saves) - render No data status when error or no url
- split widget sizes into two -- graph widget sizes and widget sizes
(graph widgets are a subset of chart widgets)
- on draft creation, do not fetch the full version. Avoid the fetch of
steps and trigger
- on activation, we were performing 8 queries/mutations synchronously +
2 additional for automated triggers. I refacto the call it it gets
reduced to 4 queries/mutations + 2 additional for automated triggers
## Context
If selected date field of the calendar view has readonly, we should not
allow users to drag/drop cards nor allow them to add a new record
(resulting in a permission error on the BE due to the FE updating the
said field)
# Introduction
We introduced a foreign key addition that will fail in production due to
orphan views targetting non existing fields
## Migration
The migration will be run for any new workspace successfully or any
twenty instance without corrupted data
## Upgrade command
The upgrade command will at some point allow the migration to be run
manually after removing any corrupted data
## Release note
We should remove the migration we've manually set as being run in
production
## Context
Switching from kanban layout to calendar was not displaying records
properly. Adding the missing state update (similar to kanban case a few
lines above)
- Allow users to choose the date granularity of the x axis and the group
by of the y axis on a bar chart
- Display those options conditionally
- Store timezones in graphs: each graph has its own timezone, defaults
to the user timezone. There will be a picker in the v2 to choose the
timezone. For now the timezone is not used by the backend, but it will
be used in filters and in group by queries.
- Store first day of the week
https://github.com/user-attachments/assets/66a5d156-dd93-4ebe-8c8f-d172f93e25be
For demo purposes, we want to work with workspaceMember has a non-system
object, allowing it to be displayed on the product, to be added custom
fields etc.
It is something we may want to do later anyway.
This PR adds a bypassPermissionChecks on workspaceMember repository
calls. This does not provide more permissions than before for other
workspaces: workspaceMember being a system object, permissions are
already bypassed for this one. (Note that actions such inviting a new
workspaceMember, updating a workspaceMember are protected by a system
permission checked in the relevant places).
This work does **not** allow any workspace to switch their
workspaceMember object to non-system, the switch is still manual.
## Description
- This PR approaches to solve
https://github.com/twentyhq/twenty/issues/15201
- updated `createDryRunResponse` to return fully populated merged record
- this way the frontend can render the populated data it as-is without
recomputing relations
- Dry-run now uses the same nested-relations population path as the
non-dryRun flow
---------
Co-authored-by: Etienne <45695613+etiennejouan@users.noreply.github.com>
fix migration command to enable the id addition in the fieldmetadata
options of workflow runs
Isues was on the workfluw rin (xurrently in produciton) if we filter by
status: clicking "Stopped" also selects "Stoppping" automatically.
<img width="735" height="436" alt="Screenshot 2025-11-03 at 12 26 40"
src="https://github.com/user-attachments/assets/20fd8b71-f7be-4115-acae-9b36f53e6d5f"
/>
In v1.8, we have already run a command to deprecate FULL or PARTIAL sync
stages.
However the code was fully deprecated in v1.10 and some workspaces might
still have this status used. This is to double check
The removal of views from the workspaces schema implies the deletion of
```
view
viewFilter
viewFilterGroup
viewGroup
viewSort
```
Before it is created again in the core schema.
However, the syncmetadata that executes the pending migration fails
because it will try to
`query failed: DROP TABLE "workspace_xxxx"."view"`
before the removal of the other view related tables that contains a view
foreign key with this kind of message
> constraint FK_c5ab40cd4debb51d588752a4857 on table "viewField" depends
on table view
Closes [Core Issue
#1772](https://github.com/twentyhq/core-team-issues/issues/1772).
<!-- CURSOR_SUMMARY -->
---
> [!NOTE]
> Introduces SSO bypass with a new permission flag and workspace-level
provider toggles, enabling permitted users to log in via
Google/Microsoft/Password when SSO-only, with backend enforcement and
frontend UI/hooks/queries.
>
> - **Backend**:
> - **Permission & Enforcement**: Add `PermissionFlagType.SSO_BYPASS`;
update `AuthService` to allow login via non-SSO providers when workspace
bypass is enabled and user has `SSO_BYPASS`.
> - **Workspace Model**: Add `isGoogleAuthBypassEnabled`,
`isMicrosoftAuthBypassEnabled`, `isPasswordAuthBypassEnabled`
(migration, entity, update input, service validation).
> - **Public API**: Extend `PublicWorkspaceDataOutput` with
`authBypassProviders`; resolver computes it; permissions defaults
include `SSO_BYPASS`.
> - **Frontend**:
> - **GraphQL/State**: Generate new types/fields; add
`authBypassProviders` to `GetPublicWorkspaceDataByDomain`; new states
`workspaceAuthBypassProvidersState`, `workspaceBypassModeState`.
> - **Auth UI/Logic**: Add `useWorkspaceBypass`; update sign-in form and
footer to offer "Bypass SSO" and use merged providers when enabled;
remove auto-redirect when single SSO.
> - **Settings**: Add Security section to toggle bypass methods per
provider; conditionally show Change Password via `useCanChangePassword`.
> - **Tests/Mocks**: Update mocks and tests to include bypass
flags/providers.
>
> <sup>Written by [Cursor
Bugbot](https://cursor.com/dashboard?tab=bugbot) for commit
8c393b2bad. This will update automatically
on new commits. Configure
[here](https://cursor.com/dashboard?tab=bugbot).</sup>
<!-- /CURSOR_SUMMARY -->
---------
Co-authored-by: Félix Malfait <felix.malfait@gmail.com>
Resolves [Dependabot Alert
255](https://github.com/twentyhq/twenty/security/dependabot/255) - fix:
tmp allows arbitrary temporary file / directory write via symbolic link
`dir` parameter.
Updated the dev-dependency `zapier-platform-cli` for it to depend on tmp
0.2.4 and also ran `yarn up tmp --recursive` to update the version of
tmp elsewhere.
Not expecting any breaking changes to twenty-zapier since
`zapier-platform-cli` is marked as a development dependency.
Co-authored-by: martmull <martmull@hotmail.fr>
# Adding MORPH support to the relationDecorator
Extends the @WorkspaceRelation decorator to support morph relations by
adding isMorphRelation and morphId options
The implementation enforces type safety through TypeScript discriminated
unions, ensuring morphId is required when isMorphRelation: true, and
includes runtime validation that throws a descriptive error if morphId
is missing.
The morph relation metadata is properly stored in metadataArgsStorage
and integrates seamlessly with the existing StandardFieldRelationFactory
to convert decorated fields into FieldMetadataType.MORPH_RELATION
entities with appropriate database constraints.
To test :
in person entity, add
```ts
@WorkspaceRelation({
standardId: PERSON_STANDARD_FIELD_IDS.testRelatedRecordCompany,
type: RelationType.MANY_TO_ONE,
label: msg`Test Related Record of Company`,
description: msg`Test relation to company`,
icon: 'IconLink',
inverseSideTarget: () => CompanyWorkspaceEntity,
inverseSideFieldKey: 'testRelatedPeople',
onDelete: RelationOnDeleteAction.CASCADE,
isMorphRelation: true,
morphId: PERSON_STANDARD_FIELD_IDS.testRelatedRecordMorphId,
})
@WorkspaceIsNullable()
@WorkspaceIsSystem()
testRelatedRecordCompany: Relation<CompanyWorkspaceEntity> | null;
@WorkspaceJoinColumn('testRelatedRecordCompany')
testRelatedRecordCompanyId: string | null;
@WorkspaceRelation({
standardId: PERSON_STANDARD_FIELD_IDS.testRelatedRecordOpportunity,
type: RelationType.MANY_TO_ONE,
label: msg`Test Related Record of Opportunity`,
description: msg`Test relation to opportunity`,
icon: 'IconLink',
inverseSideTarget: () => OpportunityWorkspaceEntity,
inverseSideFieldKey: 'testRelatedPeople',
onDelete: RelationOnDeleteAction.CASCADE,
isMorphRelation: true,
morphId: PERSON_STANDARD_FIELD_IDS.testRelatedRecordMorphId,
})
@WorkspaceIsNullable()
@WorkspaceIsSystem()
testRelatedRecordOpportunity: Relation<OpportunityWorkspaceEntity> | null;
@WorkspaceJoinColumn('testRelatedRecordOpportunity')
testRelatedRecordOpportunityId: string | null;
```
in opportunity,
```ts
@WorkspaceRelation({
standardId: OPPORTUNITY_STANDARD_FIELD_IDS.testRelatedPeople,
type: RelationType.ONE_TO_MANY,
label: msg`Test Related People`,
description: msg`People linked to the opportunity via Test Related Record`,
icon: 'IconUsers',
inverseSideTarget: () => PersonWorkspaceEntity,
inverseSideFieldKey: 'testRelatedRecordOpportunity',
onDelete: RelationOnDeleteAction.SET_NULL,
})
@WorkspaceIsNullable()
@WorkspaceIsSystem()
testRelatedPeople: Relation<PersonWorkspaceEntity[]>;
```
in company,
```ts
@WorkspaceRelation({
standardId: COMPANY_STANDARD_FIELD_IDS.testRelatedPeople,
type: RelationType.ONE_TO_MANY,
label: msg`Test Related People`,
description: msg`People linked to the company via Test Related Record`,
icon: 'IconUsers',
inverseSideTarget: () => PersonWorkspaceEntity,
inverseSideFieldKey: 'testRelatedRecordCompany',
onDelete: RelationOnDeleteAction.SET_NULL,
})
@WorkspaceIsNullable()
@WorkspaceIsSystem()
testRelatedPeople: Relation<PersonWorkspaceEntity[]>;
```
In constants/standard-field-ids.ts, add the standard fields ids
```ts
// COMPANY_STANDARD_FIELD_IDS
testRelatedPeople: '20202020-7a90-47e9-b31f-916a716fd212',
// OPPORTUNITY_STANDARD_FIELD_IDS
testRelatedPeople: '20202020-7a90-47e9-b31f-916a716fd213',
// PERSON_STANDARD_FIELD_IDS
testRelatedRecordOpportunity: '20202020-5a90-47e9-b31f-916a716fd211',
testRelatedRecordCompany: '20202020-1eb2-4298-910b-66b015b36d72',
testRelatedRecordMorphId: '20202020-6ccf-48e3-bb92-bc9961bc011e',
```
On new workspaces, as they come without the "view" object which not part
of engine metadata, favorites pointing to views make the app crashes.
This fixes it
