Data retention policy
Vybe retains Customer Data, including Slack messages, files and conversation history exchanged with Vybe AI coworkers, for as long as the customer's Vybe organization is active, so coworkers can keep the context of ongoing conversations and tasks. Slack OAuth tokens are stored encrypted and kept only while Vybe is installed in the workspace. Customer Data is never used to train AI models.
Data archiving and removal policy
Vybe removes Customer Data when it is no longer needed to provide the service. When Vybe is uninstalled from a Slack workspace, its Slack tokens and installation data are deleted immediately. When an AI coworker is deleted, its Slack app and conversation history are deleted. Customers can request deletion of all their data at any time by contacting support@vybe.build, and Vybe completes deletion within 30 days, including backups.
Data storage policy
Vybe stores Customer Data in managed cloud infrastructure: a PostgreSQL database for structured data and encrypted object storage for files. Data is encrypted in transit (TLS) and at rest. Sensitive credentials, such as Slack and integration tokens, are additionally encrypted at the application level. Access to production data is restricted to authorized Vybe personnel.
Data center location(s)
United States
Data hosting details
Cloud hosted. The application runs on Vercel, data is stored in a managed PostgreSQL database (Neon) and in cloud object storage. No on-premise infrastructure.
Data hosting company
Neon (AWS), Vercel
App/service has sub-processors
yes
Guidelines for sub-processors
App/service uses large language models (LLM)
yes
LLM model(s) used
Multi-model: Anthropic Claude, OpenAI GPT, Google Gemini/Gemma, xAI Grok, DeepSeek, Moonshot Kimi, Zhipu GLM, Alibaba Qwen and Meta models, via commercial APIs (Vercel AI Gateway). Customers pick the model per AI coworker. No training on customer data.
LLM retention settings
Commercial APIs with no training on API data. Providers may keep inputs/outputs up to 30 days only for abuse monitoring, per their terms. Conversation history is kept by Vybe per its data retention policy.
LLM data tenancy policy
Vybe hosts no LLM. Requests go via Vercel AI Gateway to the chosen provider under commercial API terms that bar training on customer data. Each request holds only that organization's data; data is never shared across customers.
LLM data residency policy
LLM requests are processed on the selected provider's infrastructure, primarily in the United States. Prompts, responses and conversation history are stored by Vybe in its own database, not by the LLM provider.