Rainbook trust centre
Rainbook is a CRM that runs inside your Jira Cloud site. Anyone Jira lets see a record sees all of it. Rainbook's AI sees what you can see, and nothing more. Your data stays in your Atlassian cloud. Your records are Jira work items, and Rainbook's indexes, history and settings are kept in Forge storage for your site. Rainbook Mail Sync connects to your Microsoft 365 or Google Workspace to bring customer email and meetings onto those records.
What is stored where
| What | Where | Who sees it |
|---|---|---|
| Deals, leads, accounts, contacts, conversations, meetings and their fields, comments and files | Jira work items in your site | Whoever Jira lets open the work item |
| Captured email: bodies, senders, recipients, files and inline images | Comments and attachments on Conversation work items in the Activity space | Whoever can open the account or deal the thread belongs to; a thread not yet linked to a record, everyone in the CRM, except on sites with Legal-pack walls, where new threads stay Private to the mailbox owner until linked |
| People index, email index, reporting tables, activity history, AI working state and settings | Forge storage for your site (Forge SQL, key-value storage and Object Store), hosted by Atlassian | Shown only through the records each person can open |
| Mailbox connections and their tokens | Rainbook Mail Sync's Forge storage, tokens as encrypted Forge secrets | Nobody: tokens are never shown to anyone, and Rainbook Mail Sync sends them only to Google or Microsoft |
| App logs: Atlassian account IDs, the cloud ID, site host and record keys | Atlassian's Forge logs, 30 days | Human Renaissance engineers, on sites that grant log access |
Rainbook stores your data in your Jira site's data residency region; AI requests run in Atlassian's published Forge LLM regions. Rainbook Mail Sync keeps its connections, sync progress and capture ledger in Forge storage and writes captured email into your Jira work items, all in your site's region; it isn't pinned to a location, because it reads mail through Google's or Microsoft's services, under your own tenant's settings. If you use Atlassian Guard data classification, classify the Activity space, where captured email lives. Rainbook runs in Atlassian's commercial cloud; it is not available on Atlassian Government Cloud.
Built on Atlassian
Rainbook runs on Atlassian-hosted infrastructure, and its only outbound connections are usage analytics and error reports, which carry no names, email addresses or record content. Its code, your records and its AI all stay on Atlassian's Forge platform. Rainbook Mail Sync is a separate app because it connects to Google and Microsoft, so it carries no Runs on Atlassian badge. The two apps share no storage; they exchange signed messages through Jira and one web endpoint.
Who sees what
- Jira decides. Every Rainbook screen, report, sidebar, Rovo answer and AI result shows a record only to people Jira lets open it. Reports and search over Rainbook's own storage are filtered by the same Jira permissions and security levels.
- You choose the audience. During setup you choose who sees customers, deals and their email: everyone in Jira, or only CRM users when you limit Rainbook to a group. With Everyone in Jira as the use setting, Only CRM users is unavailable. Deals and accounts can also be limited to their account team, and a private deal to the people it is shared with.
- Email follows its record. Everyone who can open an account or deal reads its email and meetings. Threads linked to no record are visible to your Rainbook users. Any Rainbook user who can open a thread can mark it Private, so only its owner and the people Jira allows see it. By default, that means the owner and members of rainbook-admins.
- Access changes take effect at once on records, and within an hour in team lists. Opening a record, and every private deal, private account and restricted item, is checked with Jira as the person asking, so a removal applies immediately. A person removed from an account team, or from the admins or managers group, can still see that team's Accounts in lists, boards and totals for up to an hour, until their access refreshes; opening one is refused straight away.
- Jira Free. Everyone on your Jira site works in Rainbook together. Private deals and Private threads need Jira Standard or above.
- Who uses it. Everyone who can open the CRM space in Jira uses Rainbook; your admins can limit it to a Jira group.
- Admins. Your Jira admins run Rainbook's settings. Imports, activity history, privacy requests, diagnostics, the nightly archive and Download all data stay with members of rainbook-admins, a Jira group your organisation controls. Both are checked live on every admin action. Private deals and threads open only to the people their Jira security level includes: by default, the owner and members of rainbook-admins. Being a Jira admin adds nothing: a Jira admin outside the level doesn't see the record in Rainbook, search, reports or AI.
AI and Rovo
Rainbook's AI runs on models hosted by Atlassian. It reads what you can see in Jira and nothing more. Your data isn't used to train models.
Ask Rovo about your pipeline; Claude, ChatGPT and other outside AI clients are coming soon. Each person sees exactly what Jira lets them see, and AI clients will get the same answers the Rainbook agent gives that person. Admins choose which AI features are on in Settings › AI and Rovo. The AI Usage Policy lists what each feature sends and how long results are kept.
Sensitive data
Rainbook is built for customer relationships, not for health records, government ID numbers or other sensitive personal data, and the end user terms don't allow them. Payment card numbers and US Social Security numbers are masked wherever Rainbook writes text: captured and logged email, calendar events, imports, Rainbook's forms and AI output. Each number is replaced in place with a marker such as "[card ending 4242 removed]", and the rest of the text is kept. Attached files, file names, images and text typed directly in Jira's own screens are not scanned.
Security controls
- Every request checks the caller, the licence and Jira permissions first, and refuses when a check cannot complete.
- Each Jira site's data is isolated by Atlassian's Forge platform, and a request for one site never reads another site's data: anything Rainbook keeps in memory is kept per site.
- Rainbook Mail Sync reads mail and calendars with read-only permissions, can be limited to chosen mailboxes and never sends, changes or deletes mail.
- Tokens are held only as encrypted Forge secrets in Rainbook Mail Sync. Mailbox tokens are deleted as soon as a mailbox is disconnected, and Google sign-in tokens are revoked with Google first.
- Activity history records who changed what and when, searchable for 13 months on every edition, with CSV export.
- Members of rainbook-admins can export or erase a person by email address, including people who never used Rainbook, and stop future capture of an address.
Data Handling
App logs hold Atlassian account IDs, the cloud ID, site host and record keys, and are kept for 30 days in Atlassian's Forge logs. They never hold names, email addresses, subjects, message bodies, file names, field values, prompts or AI output. Details that help with a problem stay in your own site, where admins see them: failed-capture summaries, AI results' sources for 30 days, and the diagnostics an admin can copy from Settings › Overview › Support. Read the full Data Handling document.
Encryption at Rest
Atlassian encrypts Forge storage at rest with AES-256, and Jira data at rest under Atlassian's cloud controls. All traffic between your browser, Jira, the apps, Google and Microsoft is encrypted in transit. Mailbox tokens are additionally stored as encrypted Forge secrets. Read the full Encryption at Rest document.
Incident Response
We tell affected customers about a security incident that affects their data within 72 hours of confirming it, with what happened, what data was involved and what we are doing. Critical issues get a first response within 2 hours, around the clock. Security fixes meet Atlassian's cloud deadlines: critical within 10 days, high within 4 weeks, medium within 12 weeks and low within 25 weeks. Read the full Incident Response document.
Vendor access
Human Renaissance staff have no standing access to your data. Support starts from the diagnostics an admin chooses to share from Settings › Overview › Support; message bodies are included only when the admin ticks the record they are asking about. For hands-on help, an admin can invite support@rainbook.humanr.ai as a Jira user for a limited time, and we ask for log access only when the diagnostics don't explain the problem. Staff read customer email only when an admin includes it for support, or for security or legal reasons.
Certifications and testing
Rainbook runs on Atlassian's certified cloud platform. Rainbook holds no SOC 2 or ISO 27001 certification of its own. An independent penetration test runs in January and February 2027, and its summary letter will be posted here when it is complete.
Subprocessors
| Subprocessor | What it does | What it receives | Where | When |
|---|---|---|---|---|
| Atlassian | Hosts Rainbook and Rainbook Mail Sync: Jira, Forge storage, Forge LLM and app logs | Everything the apps hold | Your Jira site's data residency region; Forge LLM in Atlassian's published regions | Always |
| Gmail and Google Calendar capture, and the Gmail add-on, which runs on Google Apps Script | Mail and calendar data of the mailboxes your organisation connects, and the email open in the add-on | Your organisation's Google Workspace | When your organisation connects Google Workspace | |
| Microsoft | Microsoft 365 mail and calendar capture, and the Outlook add-in | Mail and calendar data of the mailboxes your organisation connects, and the email open in the add-in | Your organisation's Microsoft 365 tenant | When your organisation connects Microsoft 365 |
| Cloudflare | Static website hosting, sign-in redirect pages, cookieless website analytics and the website's forms | Website visits, sign-in codes in transit and what you type in a website form; stores no customer data | Cloudflare's global network | Always |
| PostHog | Product usage analytics | Screens, features, timings and setup steps, keyed by pseudonymous site and user IDs; never names, email addresses, record content or prompts | United States (PostHog's US cloud) | On by default; admins can switch it off in Atlassian's admin controls |
| Sentry | Error reports | Error type and stack trace, keyed by a scrambled site ID only; never a person, name, email address, record content or prompts | United States (Sentry's US region; the apps refuse any other) | On by default; admins can switch it off in Atlassian's admin controls |
| Loops | Product and onboarding email | Name, email address and company of licence and evaluation contacts and of people who sign up for release updates or the launch list on the website, with the launch video and platform that sent them and the time they agreed to email | United States (Loops' servers) | Until the person unsubscribes, with one click |
| logo.dev | Company logos for new and enriched accounts | A company's web domain only | Mainly the United States (logo.dev's servers) | When Rainbook Mail Sync adds a company's logo |
| Salesforce | Import from Salesforce | The import your admin starts, on your organisation's own Salesforce account | Your organisation's Salesforce account | From January 2027, only when your admin turns it on |
| HubSpot | Import from HubSpot | The import your admin starts, on your organisation's own HubSpot account | Your organisation's HubSpot account | From January 2027, only when your admin turns it on |
| Pipedrive | Import from Pipedrive | The import your admin starts, on your organisation's own Pipedrive account | Your organisation's Pipedrive account | From January 2027, only when your admin turns it on |
| Apollo | Contact and company enrichment | A contact's email address or a company's web domain, on your organisation's own Apollo key | United States | From January 2027, only when your admin turns it on |
| People Data Labs | Contact and company enrichment | A contact's email address or a company's web domain, on your organisation's own People Data Labs key | United States | From January 2027, only when your admin turns it on |
| European Central Bank | Daily exchange rates for deal values in other currencies | A request for the day's rates; nothing about your organisation | European Union | From January 2027, only when your admin turns it on |
| SAM.gov | US federal contract opportunities for government contractors | The search filters your admin sets, on your organisation's own SAM.gov key | United States | From January 2027, only when your admin turns it on |
| Slack | Rainbook alerts in a Slack channel | The alert text your admin chooses to send, to your organisation's own Slack incoming webhook | Your organisation's Slack workspace | In a later release, only when your admin connects your organisation's own Slack webhook; nothing is sent until then |
| Microsoft Teams | Rainbook alerts in a Microsoft Teams channel | The alert text your admin chooses to send, to your organisation's own Teams workflow | Your organisation's Microsoft 365 tenant | In a later release, only when your admin connects your organisation's own Teams workflow; nothing is sent until then |
| Zapier | Rainbook events in your Zapier automations | The events and fields your admin chooses to send, to your organisation's own Zapier webhook | United States (Zapier's servers) | In a later release, only when your admin connects your organisation's own Zapier webhook; nothing is sent until then |
| Make | Rainbook events in your Make scenarios | The events and fields your admin chooses to send, to your organisation's own Make webhook | Your Make organisation's region (EU or US) | In a later release, only when your admin connects your organisation's own Make webhook; nothing is sent until then |
| Brave Search | Web search for account and contact research | The search terms, such as a company name or web domain, on your organisation's own Brave Search API key | United States | In a later release, only when your admin connects your organisation's own Brave Search API key; nothing is sent until then |
| Anthropic | AI features on your own Anthropic key, instead of Atlassian-hosted models | The prompt and the records it needs, on your organisation's own Anthropic key | United States | In a later release, only when your admin connects your organisation's own Anthropic key; nothing is sent until then |
| OpenAI | AI features on your own OpenAI key, instead of Atlassian-hosted models | The prompt and the records it needs, on your organisation's own OpenAI key | United States, or the region your OpenAI project uses | In a later release, only when your admin connects your organisation's own OpenAI key; nothing is sent until then |
| Microsoft Azure AI | AI features on your own Azure OpenAI or Azure AI Foundry resource (including Claude and other Foundry models), instead of Atlassian-hosted models | The prompt and the records it needs, on your organisation's own Azure resource and key | The Azure region your organisation chose for its resource | In a later release, only when your admin connects your organisation's own Azure key; nothing is sent until then |
We announce a new subprocessor at least 30 days before it receives data. The subprocessor list is kept current.
Data inventory
When you uninstall Rainbook Mail Sync, capture stops and Mail Sync revokes the Google sign-in tokens of people who connected their own Google mailbox, most recently used first. Atlassian deletes the app's Forge-hosted data 28 days after uninstall; a reinstall requested within 21 days can be relinked to that data. Reconnect mailboxes after a reinstall.
| Data | What it holds | Where it lives | Kept after uninstall |
|---|---|---|---|
| Contact details | A contact's email addresses, phone numbers, job title and LinkedIn URL | Fields on Contact work items | Yes |
| Lead details | A prospect's company, email address and phone number | Fields on Lead work items | Yes |
| Do not contact | A contact's wish not to be contacted, which drafts and sequences honour | A field on Contact work items | Yes |
| Teams, owners and approvers | Jira account IDs | Jira user fields | Yes |
| Deal and account values | Amounts, recurring revenue, probability and dates: business data, not personal data | Fields on deals, renewals and accounts | Yes |
| Rainbook People | The contacts on a deal or account and their buying roles | A Rainbook field, also kept in a work item property | The property, yes; the field's values only within Atlassian's reinstall window |
| Revenue impact and Health | Computed revenue and health scores | Rainbook fields, also kept in work item properties | The properties, yes |
| Captured email | Email bodies, senders, recipients, every attached file and inline image | Comments and attachments on Conversation work items | Yes |
| Message, thread, meeting and contact details | Email addresses and display names, with one-way hashes for matching | Work item and comment properties | Yes; erasing a person removes their addresses |
| Meetings | Start and end, join link, agenda, location and attendees | Fields on Meeting work items | Yes |
| People index | Email addresses and names seen on captured threads | Forge storage for your site, and the nightly Rainbook Archive work item | In the archive, yes, without erased people; the email index, no |
| AI summaries, briefs and reasons | Text written by Rainbook's AI about a record | Fields and comments on the record, and the nightly Rainbook Archive work item | Fields and comments, yes; the current ones also in the archive |
| Shared with | The Jira account IDs of the people a private deal is shared with | A Jira user field | Yes |
| Commitments | A customer's promised date and their own words | Fields on Commitment work items | Yes |
| Agent runs: who | The Jira account IDs of who started an agent run, who it worked with and who answered it, kept 400 days | Forge storage for your site, and the nightly Rainbook Archive work item | In the archive, yes |
| Agent runs: what was said | Quoted customer text and a person's answers, kept 400 days | Forge storage for your site, and the nightly Rainbook Archive work item | In the archive, yes |
| Lost deals linked to ideas | Who linked a lost deal to an idea, and the deal's reason for the loss | Forge storage for your site, and the nightly Rainbook Archive work item | In the archive, yes |
| Proposal reviews | A reviewer's reason for turning down a proposed change, up to 500 characters | Forge storage for your site | No |
| Email previews and search terms | Short message previews, sender names and keywords | Forge storage for your site, rebuilt from Jira | No |
| Teammates on threads | Jira account IDs; teammates never become Contacts | Forge storage for your site | No |
| Jira user email addresses | The email addresses of Jira users and service-desk customers, to recognise colleagues and requesters | Forge storage for your site, rebuilt from Jira | No |
| Activity history | Who changed what, with old and new values; never message bodies | Forge storage, and the nightly Rainbook Archive work item | In the archive, yes |
| AI traces | The prompt, sources and output of each AI result, kept 30 days and never used for training | Forge storage for your site | No |
| Usage | Which days each person used Rainbook, and which areas, kept 13 months | Forge storage for your site | No |
| Failed-capture summaries | Record key, subject, sender, received time, file name and error, kept until Retry or Discard and at most 30 days | Forge storage for your site | No |
| Linking rules and block lists | Email addresses on linking rules and on never-log and block lists; erased people only as one-way hashes | Forge storage for your site | In the archive, yes |
| Open privacy requests | The address and name of a person being exported or erased, replaced by hashes and counts when the request completes | Forge storage for your site | In the archive, yes |
| Company logos | A company's public logo | An attachment on the Account | Yes |
Documents
- Privacy policy
- End user terms
- Data processing agreement
- Data Retention
- Subprocessors
- AI Usage Policy
- Security and privacy answers
- CAIQ Lite answers: the Cloud Security Alliance's standard vendor questionnaire, answered for both apps, ready for your security review
- Data Handling
- Encryption at Rest
- Incident Response
- Architecture Diagram for Rainbook Mail Sync
- Service levels
- Security and vulnerability reporting
Report a vulnerability
Email security@rainbook.humanr.ai, also listed in /.well-known/security.txt. In scope: Rainbook, Rainbook Mail Sync, the Gmail add-on, the Outlook add-in and rainbook.humanr.ai. We acknowledge a report within two business days and tell you what we will fix and when.
Contact
Privacy: privacy@rainbook.humanr.ai. Security: security@rainbook.humanr.ai. Support: support@rainbook.humanr.ai, Monday to Friday, 09:00 to 17:00 New York time, except US federal holidays.