Product
Product Data Sheet
The reference specification for Kavanah — deployment model, functional modules, editions and limits, integrations, platform support, and administration. Written to be skimmed by an evaluator and attached to an RFP response.
Last updated: July 28, 2026
Kavanah is an AI-native work-management platform delivered as multi-tenant SaaS. This data sheet summarizes what the product is and what it includes. Deeper documents are linked at the end — the Technical Data Sheet for infrastructure and control detail, the Executive Product Overview for the business case.
At a glance
| Attribute | Specification |
|---|---|
| Product name | Kavanah |
| Category | Work management / project management with an integrated, governed AI agent |
| Deployment model | Multi-tenant SaaS. No self-hosted or on-premise option today. |
| Platforms | Web app; desktop app for macOS, Windows, and Linux; mobile apps for iOS and Android with native push — all built from one codebase |
| Identity | Signed-JWT sessions, bcrypt password hashing, SAML 2.0 and OIDC SSO, TOTP and WebAuthn MFA, OAuth sign-in (Google, GitHub), SCIM directory sync |
| API | Documented HTTP API with an OpenAPI 3.1 specification, 500+ endpoints; workspace API keys (prefix kvn_live_) inherit the minting user's permissions |
| AI | Anthropic Claude models via a governed tool-calling agent loop; ~896 registered agent tools; custom agent identities (AI Employees) with scopes and member access control; safe mode, action-policy approvals, action ledger, undo stack |
| Hosting | Node.js serverless functions on Vercel Fluid Compute with managed Neon Postgres; production data processed in the United States |
| Editions | Starter (free), Basic, Pro, Enterprise |
| Support contacts | Sales via /contact · security@kavanah.ai for security and procurement · support@kavanah.ai for support |
Functional modules
| Module | What it does |
|---|---|
| Work management | Projects, tasks and sub-tasks with status, assignee, due dates, dependencies, custom fields and attachments, across board, list, and timeline views. |
| Collaboration | Team chat (direct messages, group threads, channels), threaded project discussions, mentions and replies, and a notification center with email and native push. |
| Clients & portals | Client records tied to projects and work, plus scoped client-portal access so external stakeholders see live status without a workspace seat. |
| Time & billing | Time entries logged against tasks and projects, reportable by client, project, or member, with budget tracking and finance-system integration. |
| Planning & portfolio | Cross-project capacity and resourcing, sprint planning, planning horizons, goals, and estimate calibration derived from the workspace's own completion history. |
| Reporting | Productivity and portfolio dashboards, workspace usage analytics for administrators, and recurring scheduled reports delivered to people or dynamic groups. |
| AI agents / AI Employees | A governed agent that acts on workspace data through ~896 registered tools, plus named AI Employees (personas) with their own instructions, model, capability scopes, and member access control list. |
| Engineering delivery | Repositories, pull requests, deploys, incidents, and observability alerts alongside the plan, with GitHub App / OAuth repo and PR sync, and specification-to-task breakdowns. |
| Automations | Event- and time-based rules with configurable actions (for example overdue-task escalation), plus scheduled background jobs monitored by a health registry. |
| Request intake | Tokenized public intake forms that convert external submissions into tasks, with skill- and availability-aware routing to an owner. |
| Knowledge base | In-product documentation, product knowledge the agent can use to guide users to any feature, and workspace-level runbooks and memory the agent draws on. |
| Apps platform | An installable community-apps layer with its own registry and install model, so workspace-specific capability can be added without forking the product. |
A capability-by-capability breakdown, in the form procurement teams usually want for an RFP grid, is on the Capabilities Matrix.
Editions & limits
| Edition | Price | Members | Projects | Storage | Included AI credits |
|---|---|---|---|---|---|
| Starter | Free | Up to 5 | 10 | 2 GB | 250 at signup |
| Basic | $3 per seat/month ($2 per seat/month billed yearly) | Unlimited | 10 | 5 GB | 100 per seat/month |
| Pro | $6 per seat/month ($4 per seat/month billed yearly) | Unlimited | 50 | 25 GB | 250 per seat/month |
| Enterprise | Custom pricing — contact sales | Unlimited | Custom | Custom | Custom |
- AI credits beyond the included allowance are purchased separately as credit packs, at roughly $0.01 per credit.
- Yearly billing is arranged through sales rather than self-serve checkout.
- A HIPAA Business Associate Agreement is available on the Enterprise edition.
Full commercial detail is on the Pricing Sheet, and the published plans are on Pricing.
Integrations
| Category | Systems | Connection model |
|---|---|---|
| Productivity suite | Google Workspace — Gmail, Calendar, Drive, Docs, Sheets, Slides, Forms, Tasks, Meet, Chat, and Admin directory | Customer-authorized OAuth |
| Mail & calendar | Microsoft 365 / Outlook (Microsoft Graph), ProtonMail, iCloud mail and calendar | Customer-authorized OAuth or scoped credentials |
| Work & knowledge tools | Notion, ClickUp, Jira, Monday | Customer-authorized OAuth or scoped API keys |
| Finance | ClearBooks | Scoped API key |
| Engineering | GitHub — App or OAuth, with repository and pull-request sync | Customer-authorized OAuth / GitHub App install |
| Custom & programmatic | Kavanah HTTP API (OpenAPI 3.1, 500+ endpoints) and the installable apps platform | Workspace API keys (kvn_live_ prefix) inheriting the minting user's permissions |
Every integration connects through credentials the customer authorizes, is scoped to the least privilege the feature requires, and is revocable at any time from workspace settings. Stored credentials and OAuth tokens are encrypted at the application layer with AES-256-GCM under per-workspace subkeys — see the Security Overview.
Platform support
| Surface | Availability | Notes |
|---|---|---|
| Web app | All modern browsers | The primary experience, served from the deployed Next.js application. |
| Desktop app | macOS, Windows, Linux | Native shell (Tauri) over the deployed application. |
| Mobile app | iOS, Android | Native shells (Tauri) with native push notifications. |
| API / programmatic | Any HTTP client | OpenAPI 3.1 specification published at /openapi.yaml; reference at /docs. |
Because the desktop and mobile shells load the deployed application rather than a bundled copy, feature and security updates reach every platform at the same time — there is no stale client to patch and no per-platform feature gap to manage.
Administration & security
Administration
| Area | Summary |
|---|---|
| Identity & access | SAML 2.0 / OIDC SSO, TOTP and WebAuthn MFA, OAuth sign-in (Google, GitHub), SCIM directory sync, signed-JWT sessions with bcrypt password hashing. |
| Roles & permissions | Owner, admin, member, and client-portal roles, with granular capability scopes applied to administrative and AI-agent surfaces alike. |
| AI governance | Per-persona capability scopes and member access control, a safe mode that strips write tools, an action-policy engine that parks risky actions for approval, an action ledger, and an undo stack. |
| Data governance | Sensitive-field classification, GDPR/CCPA data-subject export and erasure, legal hold, and eDiscovery export. |
| Auditability | Immutable audit logs enforced by database triggers, with configurable retention (7+ years). |
Security
| Control | Summary |
|---|---|
| Encryption in transit | TLS 1.2+ with HSTS enforced on the production domain. |
| Encryption at rest | AES-256 at the database and storage layer. |
| Sensitive fields | Application-layer AES-256-GCM over integration credentials, OAuth tokens, MFA secrets, and classified PII, keyed to per-workspace subkeys derived via HKDF-SHA256. |
| Tenant isolation | Every workspace-scoped table carries a workspace_id and every query must filter on it, enforced by a continuous-integration linter that fails the build on a missing tenant filter. |
| Data residency | Production data is processed in the United States. |
Compliance status
- SOC 2 Type II— in its observation period, not yet certified.
- ISO/IEC 27001— planned, targeting Q4 2027.
- HIPAA BAA— available on Enterprise.
- Penetration testing — annual; summary available under NDA.
This section is a summary. The control-level detail belongs in the Security Overview and the Technical Data Sheet; contractual terms are in the DPA and SLA.
Where to go deeper
| Document | What it covers |
|---|---|
| Technical Data Sheet | Infrastructure, runtime, data handling, and control detail for engineering and IT reviewers. |
| Architecture Overview | The stack, application shells, API surface, integration model, and data architecture. |
| Capabilities Matrix | Capability-by-capability functional inventory, formatted for RFP grids. |
| Security Overview | Encryption, access control, tenant isolation, auditability, continuity, and secure development. |
| Implementation Guide | Provisioning, SSO and SCIM configuration, data import, integration connection, and enablement. |
| Pricing Sheet | Editions, per-seat pricing, the AI credit model, and what procurement needs for a purchase order. |
Also useful during an evaluation: Executive Product Overview, Proof of Concept, RFI/RFP pack, References, Trust & Compliance, Sub-processors, Business Continuity, AI Transparency, and the API reference.
Kavanah is in public beta. This data sheet describes product capability and current status; it is not a certification, and we do not publish customer names, adoption figures, or benchmark results. Specifications evolve — the “last updated” date above reflects the most recent revision.
Request the full evaluation pack
Procurement and security teams can request the artifacts that are not published — the SOC 2 report, penetration-test summary, and completed security questionnaires — under NDA from the Trust & Compliance page.
For pricing, a yearly agreement, or a scoped pilot, reach our team via the contact form. Product support goes to support@kavanah.ai.
Questions from a security, privacy, or procurement team? Email security@kavanah.ai. For SOC 2, penetration-test, or questionnaire artifacts shared under NDA, use the request forms on the Trust & Compliance page.