Every

Description Features Permissions Security & Compliance

@Every is an AI coworker in Slack that helps your team become AI-native. Start with a real task, see what it can do, and build from there—right where your team already works.Tag @Every in a channel or send it a direct message with a job. Connect the tools it needs, review what it produces, and give feedback in the thread. In shared channels, teammates can see what you asked for, how you guided it, and what it delivered. One person’s experiment becomes something the rest of the team can try.Use @Every to:
• Research a competitor and write a briefing for your team.
• Turn meeting notes and Slack discussions into a project plan with owners and deadlines.
• Draft and revise content for a launch.
• Investigate a code issue and open a pull request for review.
• Schedule recurring work, like a weekly metrics report or channel summary.@Every connects to tools including Google Drive, Notion, Linear, GitHub, and PostHog. You choose which channels it can access and which tools to connect.For teams with more work than people, every task is a chance to get work done and learn what to delegate next. Stop explaining AI to your team. Start showing them.Built by Every, a media and technology company sharing what it learns from running an AI-native business.Get started: https://agent.every.to
Need help? Email agent@every.to.@Every is powered by large language models and can generate inaccurate or incomplete responses. Review its work before acting on it.

Every will be able to view:

Every will be able to do:

Review the details to better understand this app’s security practices. To learn more about assessing apps for your workspace visit our Help Center.

General

Privacy & data governance

Data retention policy

@Every retains Customer Data only for as long as it is needed to provide the service. • Slack message text, and file metadata (name, type, size, and the Slack link), are stored encrypted, scoped to the installing workspace, and automatically deleted 90 days after receipt by a daily job. File contents are not held in this store. • Files a member shares for @Every to work on are retained in that member's private workspace storage so the agent can continue working with them across conversations, and are deleted on request. • Records of work performed — scheduled task definitions, run history, and audit entries — are retained for the life of the installation so members can see what @Every did on their behalf. • Conversation history sent to our AI provider is retained by that provider under our agreement with them, and is never used to train models. Slack remains the system of record. @Every stores no Slack content it does not need to answer a request or show a member what it did.

Data archiving and removal policy

When a workspace uninstalls @Every, its stored Slack credentials are deleted immediately and all processing for that workspace stops. Remaining Customer Data is deleted within 14 business days of uninstall, or of a written request to hello@every.to. That covers the message archive, stored files, task and run records, and the conversation history held by our AI provider. Requests to delete an individual member's data are honoured on the same timeline. The 90-day archive purge described above runs automatically and independently of any request.

Data storage policy

Customer Data is stored in managed cloud infrastructure in the United States. Slack message text and metadata are stored in a managed PostgreSQL database; files are stored in object storage; conversation state for the agent is held by our AI provider. Stored OAuth credentials are encrypted at rest with AES-256-GCM using versioned keys, and all traffic is encrypted in transit with TLS 1.2 or higher. Storage is scoped per workspace and per member: a request cannot read across a workspace boundary, and no internal dashboard, admin page, or support tool allows anyone at Every to browse a customer's messages.

Data center location(s)

United States

Data hosting details

Cloud hosted. @Every runs as managed services — application, background worker, and PostgreSQL database — with files in managed object storage. There is no on-premise or customer-hosted deployment.

Data hosting company

Render (application, background worker, and PostgreSQL database); Cloudflare (object storage); Anthropic (agent session state).

App/service has sub-processors

yes

Guidelines for sub-processors

https://agent.every.to/legal/privacy-policy#sub-processors

App/service uses large language models (LLM)

yes

LLM model(s) used

Anthropic Claude Opus 5 and Claude Haiku 4.5; Google Gemini 2.5 Flash (via OpenRouter)

LLM retention settings

Agent session history persists until deleted; Anthropic does not offer Zero Data Retention for it. Never used for training. Deleted on uninstall or request within 14 business days.

LLM data tenancy policy

Anthropic processes data as our processor under its Commercial Terms. Each Slack conversation is a separate agent session scoped to the installing workspace. No customer data trains any model.

LLM data residency policy

All model inference and stored conversation state is in the United States. Classifier requests are pinned to Google's US-hosted endpoints.

Certifications & compliance

Security