Skydive

Description Features Permissions Security & Compliance

Skydive lets you hire AI agents that take on real work, and gives each one its own identity in Slack. With Skydive for Slack, describe what you need and Skydive builds your own personal agents that work where your team already does. Brief them like a teammate, and watch them get to work. Skydive agents in Slack can:• Launch campaigns, handle support tickets, deliver reports, ship features, solve bugs, research and more. • Work with your entire team in channels, threads or 1:1 in DMs• Run on a schedule and report back automatically: a morning digest, a weekly rollup, or an alert the moment something changes.• Work with each other, sharing context to get bigger jobs done.• Get better automatically as they learn more about you, your company, your tools and workflows. Skydive is a manager app. When you add an agent to Slack, Skydive installs a dedicated app for the agent in your workspace, so every new agent teammate has its own identity and scoped permissions. Skydive also lets you manage your Skydive agents across the workspace.* Skydive agents are powered by AI, which is occasionally inaccurate. Review outputs before taking action.

Skydive will be able to view:

Skydive 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

Create, Inc. retains Customer Data in accordance with its Privacy Policy (https://getskydive.com/privacy) and Terms of Service (https://getskydive.com/tos). Account data and customer content (agent transcripts, memory, code, files) are retained for the life of the account, subject to in-product deletion. After account termination, exportable content is available for 60 days and customer content is deleted from active systems within 90 days. Content-level proxy logs are retained no more than 30 days; metadata, metering, and security logs up to 12 months. Backups are deleted on expiration.

Data archiving and removal policy

Customers can delete agents, transcripts, memory, files, and secrets, and disconnect connected accounts, at any time in-product or by emailing privacy@getskydive.com. Deleted content is removed from active systems within 30 days and from backups on backup expiration. On account termination, customer content is deleted from active systems within 90 days, except records we must retain for legal, billing, or security purposes. Users may also request access, correction, portability, or deletion under applicable privacy laws at privacy@skydive.com.

Data storage policy

Customer Data is stored in cloud infrastructure operated by Amazon Web Services (including S3 and RDS) in the United States, with content delivery and DNS via Cloudflare. Data is encrypted in transit and at rest. OAuth tokens and credentials are held in dedicated secrets-management infrastructure, encrypted at rest and injected at execution time so raw secrets are not exposed in the agent's working context. Data is shared only with the subprocessors needed to operate the Services (for example, AI model providers for inference) and is never sold or used for advertising. Slack and other connected-account data are never used to train AI models.

Data center location(s)

United States

Data hosting details

Cloud-hosted on Amazon Web Services (AWS), Google Cloud Platform (GCP), and Cloudflare in the United States. Agents execute in isolated cloud sandboxes (E2B; Vercel). No on-premise deployment.

Data hosting company

AWS, Google, E2B, Vercel, Cloudflare

App/service has sub-processors

yes

Guidelines for sub-processors

https://skydive.com/privacy

App/service uses large language models (LLM)

yes

LLM model(s) used

Anthropic (Claude), OpenAI (GPT), Google (Gemini), and xAI (Grok) are our primary LLM models. Some requests are routed through aggregation services (OpenRouter, Vercel AI Gateway) to user selected long tail provider.

LLM retention settings

No provider retains Customer Data for training. Anthropic, OpenAI, Google, and xAI keep inputs only briefly for abuse monitoring, then delete them; none train on it, and we never train on Slack data.

LLM data tenancy policy

Each request is processed in isolation. Our direct providers (Anthropic, OpenAI, Google, xAI) are contractually barred from training on customer content, and Slack and other connected-account data are never routed to any provider on training terms.

LLM data residency policy

Inference runs with our model providers (Anthropic, OpenAI, Google, xAI) under their API terms, primarily in the United States. Content is processed transiently to generate responses; we do not guarantee a specific region for third-party providers.

Certifications & compliance

Security