Here you’ll find the features we’ve built and are building, information around the order in which we’ll be addressing them, as well as functional and technical specifications.
We’ve mapped these features to use cases, so if you’re interested in using HASH for a particular thing, you can follow along (or even contribute!)
Knowledge Management
Data Management
Business Intelligence
Website Building
Internal Tools/Apps
Agent-Based Simulation
Entity Storage/
Inspired by the “tech trees” of video games like Civilization, we’ve broken down the key components required to build HASH into a tree-view below. You can filter these by use case to see what’s left to be built unblocking a particular use case, or slice the data by work area or completion status.
Filters
By Status
By Type
By Area
Frontend App Scaffolding
Basic application screens and frame implemented
Infrastructure
Done
UX/UI Outline
Basic application screens and functionality outline
Infrastructure
Done
Graph Layer
Rust implementation of the Graph Module atop Postgres
Infrastructure
Done
Block Protocol Graph Module
Type System and methods for accessing structured data
Infrastructure
Done
Entity Validation
Validate that entities comply with their schema when inserting and updating
User Feature
Working POC
Entity Archival (Soft Deletion)
Ability to archive or hide entities including pages
User Feature
Done
Entity Type Editor
Interface for managing entity types in HASH
User Feature
Done
Core System Types
Users, webs, and other key types required by HASH itself
Infrastructure
Done
User & Org Administration
Basic account and shared web management
User Feature
Working POC
Authorization
Permissions and access control
User Feature
Done
Advanced Authorization
More granular permissions and access control
User Feature
In Progress
Encrypted Properties
Support for selectively encrypting individual properties on entity types
Infrastructure
Future
Provisioning (LDAP)
Use an external LDAP server to provision, deprovision and sync accounts
Infrastructure
Future
LDAP Server
Use HASH as an LDAP server for authentication, authorization, and directory lookups
Infrastructure
Future
OAuth 2.0 & OIDC
Create or log in to a HASH account using a third-party service like Google
User Feature
Future
Custom OAuth Providers
Allow user/member authentication using any arbitrary OAuth provider
Infrastructure
Future
SAML SSO
Support for Single Sign On using older, legacy SAML systems
Infrastructure
Future
Sign in with HASH
Authenticate users with external services via their hash.ai accounts
Developer Facing
Done
External API & Webhooks
API endpoints for interfacing with HASH designed for external user consumption
User Feature
Future
Federated Instances
Connect a local instance of HASH to the global web of instances
User Feature
Future
Flows
Scheduled and responsive execution of user-defined logic
User Feature
Working POC
Task Executor
Temporal-based executor for flows and other logic
Infrastructure
Done
Block Protocol API Middleware
Handler for OpenAI, Mapbox and other third-party APIs
Infrastructure
Working POC
Block Protocol Service Module
Allow blocks to connect to external services without handling integration logic themselves
Infrastructure
Done
Basic Primitive Blocks
Key blocks such as heading, paragraph, image, etc.
User Feature
Done
Text Hook Provider
Hook provider for rich text-editing within blocks
Infrastructure
Done
Block Protocol Hook Module
Inject native application experiences and handlers within blocks
Infrastructure
Done
Block Protocol Core
Specification for passing structured data between blocks and embedding applications
Infrastructure
Done
@mentions
Ability to @mention users and other entities within pages
User Feature
Done
Linear Pages
Single-column block-based documents
User Feature
Done
Full-Text Search
Implementation of OpenSearch-based full text search
User Feature
Working POC
Entity Editor
Interface for managing entities in HASH
User Feature
Done
File Entity Viewer
Application-level support for rendering binary files using blocks on File entity pages
User Feature
Working POC
Realtime Service
Engine for synchronizing data between backend datastores, user sessions (via Collab), and external services (via Flows)
Infrastructure
Done
Multiplayer Collab
Live multiplayer editing and collaboration on pages (Y CRDT-based)
User Feature
Working POC
Basic Authentication
Implement sign up/in/out atop Kratos
User Feature
Done
Financial Accounting Datastore
TigerBeetle-based backend for dedicated storage and processing of financial accounting data
User Feature
Future
Brex Integration (read-only)
Ability to read information from Brex and take actions via Flows
User Feature
Future
Ramp Integration
Ability to read information from Ramp and take actions via Flows
User Feature
Future
Time Series Datastore
Dedicated backend (e.g. TimescaleDB) for optimized processing of temporal data
User Feature
Future
Similarity Search
Generation of per-entity, property, link and type embeddings, vector datastore backend, and frontend
User Feature
Done
Command Bar
Command or k-bar for quickly accessing AI capabilities and shortcuts
User Feature
Done
Shortcut Cheatsheet
Easy, centralized in-app access to keyboard navigation documentation
User Feature
Future
Keyboard Navigability
Full support for using HASH via keyboard alone
User Feature
Future
Block Protocol Actions Module
Specification for users/apps defining block action-handling
Infrastructure
Next Up
Block Action Mapping
Interface for mapping blocks to available actions
User Feature
Future
Action Blocks
Basic blocks designed for data- and action-mapping (e.g. button, dropdown, etc.)
User Feature
Future
Apps
Ability to bundle entites, types, and blocks on pages into distributable apps
User Feature
Future
App Store
Publish and distribute or discover user-created HASH Applications
User Feature
Future
Semantic Q&A
Natural language questions and answers based on the contents of a web
User Feature
Working POC
Intelligent Global Search
Augment semantic Q&A with relevant and timely context from the global web
User Feature
Working POC
Integrations: Realtime sync
Ability to synchronize information to/fro external services via the Realtime Service
Developer Facing
Done
Rippling Integration
Two-way sync of information and actions with the Rippling platform
User Feature
Future
GitHub Integration
Two-way sync of information with GitHub and ability to trigger actions via Flows
User Feature
Next Up
Git Datastore
libgit2 or Gitea-based backend for dedicated storage and processing of source code and Git repositories
Infrastructure
Future
REPL Block
Support for individually executable blocks (JavaScript/Python notebook-style) within pages
Developer Facing
Future
Optimized Simulation Run Storage
Move to a more efficient data format for storing simulation data (e.g. Parquet)
Infrastructure
Future
Improved scalability
Intelligent offloading of infrequently accessed data into cold storage
Infrastructure
Future
Self-service hCloud
Offered up until May 2023, we plan to re-introduce this as part of Flows
User Feature
Future
Simulation Blocks
Replaces hCore, enabling simulations and experiments to be created and run in HASH
User Feature
Future
Upgraded Geospatial-3D
Improved styles, performance, and combining the separate geospatial and 3D viewers in one
Infrastructure
Future
Asana Integration
Two-way sync of information with one or more Asana organizations
User Feature
Future
Linear Integration
Two-way sync of information with one or more Linear workspaces
User Feature
In Progress
Integrations: One-time write
Ability to write information out to connected external services
Developer Facing
Done
Integrations: One-time load
Ability to ingest information from external services as entities
Developer Facing
Done
Notion Integration
Ability to one-time import information from Notion (two-way sync when Notion supports webhooks)
User Feature
Future
Block-Level Comments
The ability to comment on blocks within pages
User Feature
Done
Inline Comments
Comment precisely (e.g. on a span of text, or part of an image) inside blocks
User Feature
Future
Canvas Pages
Freeform drag-and-drop canvases for blocks
User Feature
Working POC
Mobile App
iOS & Android applications, with full offline mode
User Feature
Future
Desktop App
macOS, Windows & Linux applications, with full offline mode
User Feature
Future
One-click self-hosting
Host your own HASH instance with no technical setup required
User Feature
Future
Canvas AutoLayout
Dynamically position blocks relative to one another
User Feature
Future
Inbox (Notifications)
User-facing in-app alerts upon notifiable events
User Feature
Working POC
Browser Extension
Find, use and create entities as you browse
User Feature
Done
Draft New Entities
New suggested, incorrect or incomplete entities may be saved as drafts
User Feature
Done
Draft Updated Entities
Updates or suggested changes to entities may be persisted as drafts prior to finalization
User Feature
Done
AI Entity Creation
LLM-assisted new entity creation
User Feature
Done
AI Type Creation
LLM-assisted new and existing type suggestions
User Feature
Working POC
Data Querying & Selection
Interface for identifying and loading entities into blocks
User Feature
Working POC
Advanced Blocks
Blocks which allow users to query and insert multiple entities of varying types (e.g. table, kanban, timeline)
User Feature
Working POC
Composite Blocks
Nest blocks inside of one another to create more powerful user experiences
Developer Facing
Future
API-Based Blocks
OpenAI and Mapbox-enabled blocks built atop the Þ API middleware
User Feature
Done
Email Sending
Support for user composition and triggering of emails via Flows
User Feature
Future
Gmail/Google email sync
Read, send and manage emails with Gmail and Google Workspace
User Feature
Future
Type Archival (Soft Deletion)
Ability to hide outdated or redundant types in the UI
User Feature
Done
Type Renaming & Redirection
Rename types already in-use, and seamlessly redirect old consumers
User Feature
Future
API Transactions
Complex API requests composed of multiple interdependent operations
Infrastructure
Future
Entity/Property Hard Deletion
Ability to permanently delete information from the backend datastore(s)
User Feature
Future
Entity Duplication
Easy duplication of existing entities with lineage information preserved
User Feature
Future
Type Inheritance
Set parent types whose expected property and link types will be inherited
User Feature
Done
Entity Labels
Select properties that can be used to easily identify entities in HASH
User Feature
Done
Data Type Groups
Ability to group data types and reference multiple at once
Infrastructure
Next Up
Custom Data Types
Interface allowing user definition of non-primitive Data Types
User Feature
Working POC
Calculated Properties
Programmatic calculation of properties, supported independently of Flows.
User Feature
Working POC
Data Type Conversion
Ability to automatically convert between data types within a group
User Feature
Future
Semantic Similarity Mapping
Specify that different types are semantically related to one another
User Feature
Future
Literal Sameness Mapping
Specify that different web entities refer to the same literal thing, and watch for changes
User Feature
Future
Type Forking
Duplicate types while retaining a reference to the original (distinct from extension)
User Feature
Future
Multi-Type Entities
Allow a single entity to have more than one assigned type
User Feature
In Progress
Quick Notes
Support for one-click capturing reverse-chronological notes
User Feature
Working POC