| Release | BloodHound | OpenHound | SharpHound | AzureHound |
| 2026-04-13 | v9.0.0 | v0.1.0 | v2.12.0 | v2.12.0 |
OpenHound Data Collector Framework
Collect and convert data from GitHub, Jamf, and Okta with OpenHound, a standardized framework for building and running OpenGraph collectors and converters.OpenHound is built in Python and powered by the Data Load Tool (DLT) library so you can follow a consistent workflow to collect, process, and convert data from external sources into BloodHound-compatible graphs. You can run OpenHound as a containerized service or as a standalone CLI application.OpenGraph Extension Management
Use the new OpenGraph Management page to upload, view, update, and delete OpenGraph extensions so you can manage structured graph support.OpenGraph extensions define the schema for your custom graph data, including node and edge types, icons, properties, and relationships, so ingestion stays consistent and predictable.This release also exposes experimental OpenGraph Management APIs so you can evaluate and automate extension workflows:| Method | Endpoint | Description |
|---|---|---|
GET | /api/v2/extensions | Get a list of all OpenGraph extensions. |
PUT | /api/v2/extensions | Upserts the OpenGraph extension. |
DELETE | /api/v2/extensions/{extension_id} | Delete an OpenGraph extension. |
GET | /api/v2/extension-edges | Get a list of all edge kinds across OpenGraph schemas. |
OpenGraph Findings
This is a SpecterOps-managed feature. If it is not enabled in your environment, contact your account team for assistance.
Environment Targeted Access Control
This is a SpecterOps-managed feature. If it is not enabled in your environment, contact your account team for assistance.
SharpHound Integrated Windows Authentication
Graph Readability Improvements
Explore graphs with improved node labels, clearer directional arrows, and more intuitive selection behavior:- Node labels now display below nodes and include a second line identifying the source platform and type (for example, Active Directory | Group). Long labels are truncated with an ellipsis and show the full text on selection.
- Edges now feature higher-contrast directional arrows that connect directly to nodes instead of node labels, making it easier to follow relationship paths at default zoom levels.
-
The Entity panel now closes when clicking on the graph background to clear selection, giving you more space to explore the graph without manually closing the panel.

Table View Column Organization
Work through table data faster with better control over column layout.This release adds stronger column organization options for table-heavy analysis tasks. You can now drag-and-drop columns to reorder them.By default, the Node Type and Name columns are pinned. Moving an unpinned column into a group of pinned columns pins it. Similarly, moving a pinned column out of a pinned group unpins it. You can manage pinned columns and reset column order and size from the Columns menu.
Improved Graph Export Filenames
Export graph data with more useful filenames to avoid repeatedbh-graph.json collisions in your local filesystem.Exported graph files now include a timestamp in the filename to make it easier to manage multiple exports without manually renaming files. For example:Navigation Sidebar Modernization
Use a more predictable navigation experience with improved expand and collapse behavior in the primary sidebar.The updated navigation sidebar defaults to expanded for new users, supports persistent collapse state, and improves discoverability of menu items and submenus.
Azure Hybrid Post-Processing and Data Hygiene
Keep hybrid Azure and Active Directory (AD) relationship data cleaner with less unnecessary graph churn during post-processing.This release optimizes role-assignment processing and stops creating placeholder AD objects during Azure ingest and hybrid post-processing so relationships are only created when both Azure and AD context is available. This significantly improves analysis time for tenants with large Azure environments.OpenGraph Node Type Filtering in Search
Filter search results by node type to find relevant OpenGraph data faster.This enhancement adds support for node type filtering for OpenGraph nodes.HasSession Edge Deletion
Refresh time-sensitive session data without clearing the entire graph by deleting only HasSession edges from the database.This enhancement adds a dedicated HasSession checkbox to the Database Management page so you can remove stale session relationships before recollecting updated session data.
Accessibility Improvements
Navigate BloodHound more effectively with assistive technologies through clearer component naming and better non-text content labeling across the interface.This accessibility work strengthens screen reader compatibility by improving role, state, and value semantics and adding or refining labels for visual UI elements.Certification Selection Behavior

Posture Metric Abbreviation
- This applies to the Count and Change columns in the Attack Paths table, and Attack Paths, Findings, and Tier Zero Objects counts in the Attack Path Summary panel.
-
After values exceed five digits, BloodHound abbreviates values using standard rounding with compact suffixes (for example,
xxxK,y.yyyM, andz.zzzB).
Custom Range Validation

API and Auth
- Resolved an issue where fuzzy search on
GET /api/v2/clientsdid not consistently apply acrossname,hostname, andconfigured_userfields. - Expanded auditor access to key collection and administration read endpoints, including completed jobs, events, file uploads, and
bloodhound-users. This release also introduced a newCollectionReadJobspermission and removed an unused application configuration management permission. - Strengthened authorization checks for the
/api/v2/tokensendpoint so low-privileged users cannot enumerate token metadata for other users through filter manipulation. - Resolved an issue where the
/api/v2/graphs/edge-compositionendpoint could return duplicate edges in ESC edge composition results. - Resolved an issue where the API Explorer did not correctly validate operator-prefixed values for query parameters, such as
eqandneq.
Cypher and Explore
- Resolved an issue where pressing the space bar in the Cypher query editor on the Explore page caused unexpected behavior.
- Resolved an issue where the Cypher autocomplete dropdown on the Explore page rendered in the wrong position and may not be visible.
- Resolved an issue where several prebuilt “Shortest paths…” Saved Queries queries could hit the default query timeout and fail without returning a graceful timeout response.
- Resolved an issue where Cypher queries on PostgreSQL using multi-hop graph traversals with variable-length path patterns (for example,
*1..) could fail, particularly for edges with traversal limits. - Resolved an issue where
OPTIONAL MATCHclauses could be translated without a valid prior query frame, causing some Cypher queries to return incorrect results.
OpenGraph
Resolved an issue causing excessive memory use during ingestion of large OpenGraph data payloads.Privilege Zones and Posture
- Resolved an issue where long Privilege Zone rule names without spaces could overflow and stretch layout containers.
- Resolved an issue preventing Affected Objects accordions for GPO objects from expanding on the Zone Builder page.
- Resolved an issue where Large Default Group findings could appear incorrectly under the Tier Zero filter on the Posture page.
- Resolved an issue where AzureHound filtered
AZResourceGroupUserAccessAdminresults by the Owner role instead of the User Access Administrator role, causing incorrectAZUserAccessAdministratoredges onAZResourceGroupnodes. - Resolved an issue where
AZContributoredges were not collected for Management Groups, Resource Groups, and Subscriptions, causing contributor role assignments to be absent from the graph for these resource types.