Skip to main content

SOC Prime Platform Product Release Notes 6.3.3

N
Written by Nataliia Pukaliak

November 18, 2026

© 2026 SOC Prime Inc.

All rights reserved. This product and documentation related are protected by copyright and distributed under licenses restricting their use, copying, distribution, and decompilation. No part of this product or documentation related may be reproduced in any form or by any means without the prior written authorization of SOC Prime. While every precaution has been taken in the preparation of this book, SOC Prime assumes no responsibility for errors or omissions. This publication and features described herein are subject to change without notice.

LogTotal


LogTotal now lets you analyze your log files. It’s available in two editions: the public LogTotal and LogTotal in Prime Hunt.

Beyond sanitization, LogTotal now evaluates the events in your file against SOC Prime detection rules. Events with rule hits are then correlated using Higher Order Sigma rules associated with Active Threats. These rules describe multi-event attack patterns. When enough of a pattern is matched, an Attack Chain is formed based on the corresponding Active Threat. Our AI model then analyzes each chain and filters out low-confidence results, so only high-confidence chains are shown by default.

Analysis results in LogTotal are organized into the following:

Attack Chains

The Attack Chains tab lists the Attack Chains formed from the rule hits. Use the available controls to filter the visibility of Attack Chains:

  • Show Filtered Out – Shows or hides chains the AI model marked as false positives or low priority

  • Group by – Groups chains by Hostname or Type

  • All Severities and All Statuses – Filters chains by severity and investigation status. Statuses available only for Prime Hunt LogTotal (statuses are available only for LogTotal in Prime Hunt)

Select an Attack Chain name to open its details page, which includes:

  • Status – The lifecycle stage of the Attack Chain

  • AI Summary – A contextual overview of the detected activity

  • Tags – MITRE ATT&CK techniques and sub-techniques, and related threat actors

  • Active Threat – The Active Threat that the attack chain is based on. Select its tile to open the Active Threat details page

  • Chain – The sequence of detected techniques over time

  • Techniques – The list of detected techniques and its rules

Rule Hits

The Rule Hits tab lists the rules that had hits on the events in your file.

Stealth Techniques and Suspicious Findings

The Stealth Techniques and Suspicious Findings tabs highlight activity that deserves a closer look: techniques from the Stealth tactic and techniques whose rules have an unusually high number of hits.

Incident Report

Once Event Analysis is complete, the platform uses AI to turn the results into a structured incident report. The report summarizes the attack chain findings, individual rule hits, and other relevant information from your uploaded file. It gives you clear containment and response measures, along with concise lessons learned.

You can copy the report to your clipboard or download it as a Markdown file.

In public LogTotal, you can select the language of the incident report from the dropdown menu before you start the analysis.

Details

The Details tab summarizes the analyzed file, including its basic properties, log period, and hashes.

Tagged Logs

The Tagged Logs page lets you explore raw events from your file. It opens in two ways:

  • From the Tagged Logs tab of an Attack Chain, where it shows the events used to build the chain

  • From the View Tagged Logs button next to a rule on the Rule Hits, Stealth Techniques, and Suspicious Findings tabs, and in the list of rules on the General tab of Attack Chain Details

The page includes:

  • Search bar – Search within the displayed events using Lucene query syntax. For example, you can search by field name and value to narrow the results to the events you need

  • Date and time picker – Narrow the selection to a specific time range (LogTotal in Prime Hunt only)

  • Export – Click Export to download the displayed events in a CSV file

  • Timeline – The number of events hits over time. At the top, the total number of hits and the time period of the current selection (LogTotal in Prime Hunt only)

  • Columns – Choose which columns appear in the events table

  • Events list – Lists each event with the following columns:

    • Timestamp – When the event occurred

    • Value – A preview of the event’s raw content, shown as JSON

    Click an event to open the full event data. Use the toggle to switch between two formats:

    • Default – A table of field names and values

    • JSON – The raw event in JSON format

Event Analysis History (LogTotal in Prime Hunt only)

Previously run analyses are listed on the Event Analysis page in Prime Hunt, so you can return to past results at any time.

API Access (LogTotal in Prime Hunt only)

You can also run Event Analysis via API. To get started, generate a personal API key in Platform Settings > API and select Prime Hunt as one of the product APIs the key can access. For endpoints, parameters, and request examples, refer to Event Analyses API.

WHY

LogTotal helps you quickly check standalone log files for signs of an attack. Instead of reviewing raw events manually, you get correlated attack chains, AI-prioritized findings, and the exact events behind them.

Prime Core


Evergreen Threats

With this release, we’ve introduced Evergreen Threats, alongside the existing Active Threats. Evergreen Threats cover long-standing threats that stay relevant over time.

As part of this update, the Active Threats page now has two tabs:

  • News – The Active Threats you already know, covering current and emerging threat activity

  • Evergreen – The new Evergreen Threats, covering long-standing threats that remain relevant regardless of recent news

The Evergreen Threat details page displays the following information about the threat:

  • Threat overview – Threat’s severity, number of views, publication date

  • Share – Select the Share button to share the news item

  • Tabs

    • AI Summary

    • Higher Order Sigma

    • Detections (behavior rules only)

  • Threat attributes – Threat Type, Affected Industries, Affected Geographies, Sub-Techniques, CVE

WHY

Evergreen Threats provide a dedicated way to explore long-standing threats that remain relevant over time, regardless of current news or emerging activity.

Company Tags

We’ve added Company Tags, which let you assign company-specific labels to detection content in Platform Repos. Tags are shared across all users in your company, with up to 100 unique tags per company.

To add a tag, select Add Tag on the Detection Rule page, then select an existing tag or enter a name and select Create New Tag.

On the Search page, use the new Company Tags filter to find detection content assigned to specific tags.

WHY

Tags let your team organize detection content around its own priorities, such as projects, use cases, or log sources, and quickly find the content it needs.

Access to Detections from Active Threats

We’ve improved how you interact with detections in the Detections section of an Active Threat News Item page in Prime Core.

  • Open Detection Details – Clicking a detection tile now opens the Detection Rule page in a new browser tab, with the Detection Intelligence tab selected. You can explore intelligence related to the threat and review the rule’s metadata.

  • View Code – A new button appears when hovering over a detection tile. Clicking it opens the Detection Rule page in a new browser tab, with the Detection Code tab selected. You can view the detection code, select an available translation, customize the rule, or deploy it.

  • Open in Architect – A new button opens the detection in the Translation workspace in Prime Architect, with the detection code added as the source content. From there, you can translate, validate, optimize, and manage detection content across multiple platforms and languages.

WHY

Working with detections on an Active Threat News Item page used to take extra steps to reach the rule details, code, or translation tools. To make this workflow faster and more intuitive, we’ve exposed the most common actions directly from the list.

Redesign of the Search Page

This release introduces a redesigned Search page in Prime Core with a cleaner, more streamlined interface that makes it easier to find and manage relevant content.

Alongside the visual refresh, we’ve also significantly improved its performance. Search results now load 34% faster on average, so everyday searches feel noticeably quicker. The biggest gains are in the searches that previously took the longest, which are now 49% faster.

WHY

Search is one of the most frequently used modules in Prime Core, so we wanted to make it as fast and efficient as possible. With the redesign, search is now faster, more consistent, and easier to use, helping you find and manage content with less waiting.

Prime Architect


Agentic Skills

With this release, we've introduced Agentic Skills in Agentic Threat Research. Skills are sets of instructions that extend the Prime Architect agent for specific tasks. A skill already includes the steps and requirements for a workflow, and the agent follows them to complete the task.

You can use library skills, which are reviewed by the SOC Prime team and published by SOC Prime or other publishers, or create your own personal skills.

Agentic Skills is available at Prime Architect > Agentic Threat Research > Agentic Skills and includes the following tabs:

  • My – Skills installed from the library and personal skills created by you

  • Bookmarks – Bookmarked skills that you want to evaluate or revisit later

  • Browse – All available library skills, which you can search by name, description, author's name, or filter by NIST Cybersecurity Framework category (Govern, Identify, Protect, Detect, Respond, Recover)

The Skill Details page includes two tabs:

  • Overview – A short description, a flow diagram of the expected input, process, and output, and expandable sections on when to use the skill, its prerequisites, and how it works, plus an example prompt. The right side shows the skill's Security Score, security check results, number of installations, release date, author, SOC Prime verification status, and compatibility with environments and agents such as Prime Architect, Claude, and Codex.

  • Content – The skill's full instructions in read-only mode, so you can see exactly what it will instruct the agent to do before installing, bookmarking, or downloading it.

Actions with skills:

  • Create a personal skill – Select Add Skill on the My tab to create a skill with a name, description, instructions, and category. Personal skills can be edited, deleted, or installed.

  • Install a skill – Install a skill to make it available for use in Prime Architect chat. Installed skills receive published updates automatically. You can install a skill in any of the following ways:

    • On the Skill Details page of a library skill, select Install.

    • On a library skill tile in the Browse tab, select Install to My Skills.

    • For a personal skill in the My tab, select the three dots on the skill tile and select Install.

  • Download a skill – On the Skill Details page, select Download to save any library or personal skill as a ZIP package containing its SKILL.md file. You can store it or use it with a compatible external agent or platform.

Use a Skill in Chat

Once a skill is installed, add it to message in one of the following ways:

  • Click the + icon in the chat input and select Add Skills

  • Type / in the chat input and select a skill from the list

  • Select Agentic Skills above the chat input and choose a skill

You can combine a skill with your own prompt text, a file, or a detection in the same message. One skill can be used per message.

WHY

Agentic Skills let you hand the agent a ready-made workflow instead of re-explaining it in every prompt, which saves time and keeps results consistent across common tasks like writing rules, building hunting queries, and extracting IOCs. They also let you reuse reviewed library skills or capture your own team's processes as personal skills.

MCP Skills Tools

Prime Architect now offers skills tools through MCP, allowing you to work with Agentic Skills from the SOC Prime library directly in their AI environment. You can have your AI agent discover skills by browsing or searching the library and your personal skills, and inspect a skill’s details and instructions before using it. A skill can then be used in your own environment or run in Prime Architect, with the results returned to your AI environment so you can continue working there.

Available tools:

list_skills – Browse library skills, filter by category, and sort by newest, most installed, or recently updated

search_skills – Search library skills by ID, name, description, or author

list_user_skills – Browse your personal skills, sorted by newest or recently updated

search_user_skills – Search your personal skills by ID, name, or description

get_skill – View the full details and instructions of a library skill

get_user_skill – View the full details and instructions of personal skill

run_architect – Run a skill in Prime Architect and receive the results directly in your AI environment

To learn more, see the MCP user guide.

WHY

You can find, review, and apply Agentic Skills together with your own tools in your AI environment, without switching to Prime Architect.

URL Support in Chat Messages

You can now add URLs, such as links to threat reports, blog posts, or CERT articles, to your chat messages to give the AI agent additional context. For example, ask the AI agent to summarize a threat report, identify MITRE ATT&CK techniques, or generate a detection rule based on the provided content.

To use this feature, paste one or more URLs into the chat input and describe what the AI agent should do with the content, or select one of the available agentic AI tools.

You can add up to 5 URLs per message. If you include more, only the first 5 are processed, so send any additional URLs in a separate message.

When processing URLs, the system confirms each successful content extraction. If a URL is unreachable, blocked, or contains no usable content, the system displays a message and continues with the remaining URLs, and the AI agent generates its response from the content that was successfully extracted. To include unavailable content in the analysis, copy and paste the page content directly into the chat.

WHY

Adding a URL as additional context for the AI agent lets you go from a published threat report or article to actionable insights faster, without manually copying and pasting its content into the chat.

Prime Hunt


Support for SentinelOne

We've added support for SentinelOne Data Planes. To run investigation in your SentinelOne instance, first configure a Data Plane integration in Platform Settings > Data Planes.

  1. Click Add Data Plane.

  2. Name your profile, select SentinelOne as your platform, and choose if you want to share the profile with your teammates.

  3. Ensure the Prime Hunt checkbox is enabled.

  4. Fill in the fields that appeared in the Configuration section:

    • API Key

    • API URL

    • Site IDs

    • Account IDS (optional)

    See this article to learn how to get credentials in SentinelOne.

  5. Click Save Changes.

Once you've set up the integration with your Data Plane, start an investigation in Prime Hunt using your Data Plane.

WHY

We’ve extended Prime Hunt to SentinelOne, so teams using this platform can automate threat searches across their data.

Agentic Skills in Active Threats


Active Threat News Items page now includes a new Agentic Skills tab showing agentic skills relevant to the threat you’re viewing.

Skills are grouped according to the NIST Cybersecurity Framework: Govern, Identify, Protect, Detect, Respond, and Recover. Switch between categories to see the skills linked to the Active Threat News item.

Viewing skill details

Click a skill tile to open its details page in Prime Architect.

Bookmarking and installing skills

You can manage skills directly from their tiles:

  • Bookmark – Click the bookmark button to save a skill for future reference, and click it again to remove the bookmark

  • Install – Click the + button to install a skill and make it available in Prime Architect chat, and click it again to uninstall it

WHY

Relevant skills appear directly on Active Threat News Items, so you can move from reading about a threat to acting on it without searching the full skill library.

SSO Group Mapping


We’ve added support for SAML group mapping, which allows a Manager to map identity provider groups to SOC Prime Platform roles and set a priority for each mapping. After the mappings are saved, the platform assigns the mapped role when a user signs in through SSO, so role assignment follows the group structure in your identity provider.

To configure group mapping:

  1. In the SOC Prime Platform, go to Account icon > Platform Settings > SSO Configuration.

  2. Click Add Mapping.

  3. Specify the following:

    • SAML group name – The full group name as configured in the identity provider. Matching is case-insensitive

    • SOC Prime platform role – The role to assign to users in the group

    • Group priority – The mapping priority, where 1 is the highest priority

  4. Click Next Step, then Save Changes to activate the mappings.

The platform applies group mappings when users with the company default role sign in through SSO. If a user belongs to multiple mapped groups, the role associated with the highest-priority mapping is assigned.

For setup instructions for Okta and Microsoft Entra ID, see the SSO Configuration guide.

Note: Group mapping doesn’t continuously sync roles with your identity provider. Once a mapped role is assigned, later group changes don’t update it. To apply mapping again, reassign the company default role to the user.

WHY

Group mapping removes the need to assign roles to SSO users manually. New users automatically get the right role on their first sign-in, based on your setup and the groups you already manage in your identity provider.

GitHub Enterprise Server Support


The GitHub integration now supports GitHub Enterprise Server. To connect a self-hosted GitHub instance, go to Account > Platform Settings > Integrations, add or edit a GitHub integration, select the GitHub Enterprise Server checkbox in the Configuration section, and enter the URL of your GitHub instance in the field that appears.

WHY

Allows organizations using self-hosted GitHub Enterprise Server to configure the GitHub integration and push detection content to their repositories as part of their automation flow.

Content Quality Improvements


CrowdStrike Next-Gen SIEM

Improved translation from Sigma to CrowdStrike Next-Gen SIEM by removing additional escaping in values.

Google SecOps

Improved translations to Google SecOps Query (UDM) and Rule (YARA-L) to address translation issues.

Microsoft Defender for Endpoint

Fixed an issue where the second part of Microsoft Defender queries was not included in the Sigma translation.

Key Bug Fixes and Improvements


  • Fixed an issue where sometimes Custom Field Mapping was not applied when using API.

  • Website updates:

    • Added LogTotal landing page to the website.

    • Updated the information on Prime Detect landing page.

    • Updated breadcrumbs to display category names and full article titles.

    • Added a new category How to Logtotal to the blog on the website.

    • Improved tables in blog articles for mobile version.

    • Updated the Attack Detective page on the website with the current Prime Hunt branding and updated product information.

    • Updated Prime Hunt landing page.

    • Updated the Uncoder AI page on the website with the current Prime Architect branding and updated product information.

    • Added Polish language support to the website.

    • Updated social media icons in blog articles to allow users to share articles on social media.

    • Added Mandarin Chinese language support for blog articles and Active Threats.

  • Fixed an issue where sometimes Splunk alerts without a name caused API requests to retrieve content from Content Lists to fail.

  • Improvements related to Prime Architect:

    • Fixed an issue sometimes causing the chat options menu in Prime Architect to shift when resizing the screen.

    • Improved handling of long text chat inputs by automatically converting text exceeding 3,000 characters into a file attachment.

    • Improved the chat interface by hiding the empty message bar when a user sends a text attachment without additional text, AI tool, or skill input.

    • Improved the display of personal skills description in Agentic Skills to prevent text overlap.

    • Fixed an issue where sometimes the wrong author name was displayed on the Skill Details page.

    • Updated the skill installation count display to show the number of installs in the {count} Installs format instead of an icon and numeric count.

    • Fixed an issue where sometimes the page crashed when translating Sigma rules to OpenTide.

  • Fixed several issues on the Log Source Coverage page:

    • Fixed an issue where an internal server error sometimes occurred on the Log Source Coverage page when selecting NPCore ZombieZero XDR Search Profile.

    • Fixed an issue where sometimes the rule counts on the Log Source page and the Search page did not match after drilling down, either when no filters were applied or when a search profile was selected.

  • Fixed an issue where an empty Data Plane dropdown was displayed when creating a Custom Field Mapping profile for a platform with no configured Data Planes.

  • Updated the link on the Data Plane Creation/Editing page to point to the correct Prime Core API Integration Tool for Elastic Stack.

  • Fixed an issue where a Log Sources Not Found error was displayed in Prime Hunt during a threat scan in Google SecOps.

  • Fixed an issue where Elastic Stack inventory jobs could get stuck.

  • Fixed an issue where the Splunk on-prem Data Plane was sometimes incorrectly displayed as Disconnected while long-running searches were still running and scans completed successfully.

  • Improved the Check Connection feature for Elastic Stack.

  • Increased the file size limit for LogTotal in Prime Hunt to 4,8 GB and the event limit to 10 million.

  • Fixed an issue where some Active Threats API endpoints were not working.

  • Updated SOC Prime Terms of Service.

Did this answer your question?