Legal Tech Release Notes
Release notes for legal practice management, case management and legal AI tools
Products (9)
Latest Legal Tech Updates
- April 2026
- No date parsed from source.
- First seen by Releasebot:Apr 17, 2026
Apr 18, 2026
GC AI introduces the Skill Library, replacing prompts with attachable skills that can be chained together for richer, cleaner instructions. It also adds Skill Creator, in-chat skill editing, and web improvements like knowledge base files, skill-aware prompt help, and a /prompts to /skills redirect.
Skills: A Better Way to Reuse Your Best Instructions
The Prompt Library is now the Skill Library, and skills work differently than prompts did. Instead of pasting instructions into the composer as plain text, you attach skills to a conversation the same way you attach files. GC AI reads the skill’s instructions as dedicated context while your actual question stays short and focused.
The biggest change: you can now chain multiple skills in a single conversation. Attach a compliance checklist skill and a plain-language drafting skill at the same time, and GC AI follows both sets of instructions together. Advanced, multi-step instructions that used to crowd the composer now live cleanly in attached skills.
All your existing saved prompts have been carried over as skills automatically.
Skill Creator
A new GC AI skill that helps you build reusable skills through conversation. Attach the “Create Skill” skill, describe what you want, and GC AI interviews you about your goal, drafts the skill, and saves it directly to your library. It knows six common patterns (checklist, role, template, workflow, style, and drafting) and coaches you through picking the right structure. You can also ask it to refine existing skills.
In-Chat Skill Editing
When GC AI suggests changes to one of your skills during a conversation, you see a side-by-side comparison of the original and proposed versions before anything is saved. Approve the changes as-is, switch into edit mode to adjust them, or deny if they miss the mark.
Improvements
Web
- Skills can now carry knowledge base files. Link files to a skill, and they travel with it every time it’s used. No re-uploading templates, policies, or example documents.
- Easy Prompt now includes attached skill context when optimizing your question, so suggestions are tailored to the skills you’re working with.
- Clicking a skill badge on a sent message opens the skill in the sidebar for quick viewing, editing, or sharing.
- The /prompts URL now permanently redirects to /skills. Existing bookmarks and shared links continue to work.
- Apr 16, 2026
- Date parsed from source:Apr 16, 2026
- First seen by Releasebot:Apr 17, 2026
Opus 4.7, Now Live in :Harvey:
Harvey adds Claude Opus 4.7, bringing stronger reasoning calibration, better accuracy on complex legal workflows, and top BigLaw Bench results for deal management, risk assessment, drafting, and document-heavy transactional work.
Today, we are making Claude Opus 4.7 available in Harvey. Opus 4.7 is Anthropic's latest frontier model, building on the agentic strengths of Opus 4.6 while delivering improved reasoning calibration, stronger performance on ambiguous tasks, and gains in substantive accuracy across complex legal workflows.
Early access evaluations point to a model that is better at matching its reasoning depth to the complexity of the task at hand. On our BigLaw Bench evaluation suite, Opus 4.7 scored 90.9%, the highest score of any Claude model in Harvey to date — 45% of tasks received perfect scores, with 88% scoring at or above 0.80. We expect Opus 4.7 to excel in deal management, risk assessment, drafting, and document-heavy transactional work where precision and analytical depth are essential.
Our teams highlighted several additional strengths. The model demonstrates notably improved reasoning calibration. For straightforward questions like identifying governing law, it returns direct answers rather than multi-paragraph explanations. On harder analytical tasks, such as distinguishing between assignment provisions and change-of-control provisions, the model performed correctly where prior models have historically struggled.
“Claude Opus 4.7 scored 90.9% on Harvey's BigLaw Bench, the highest score of any Claude model to date. Beyond raw accuracy, the model shows meaningfully better reasoning calibration: it adapts its depth to the complexity of the question, returning concise answers where appropriate and detailed analysis where it matters.”
Niko Grupen
Head of Applied Research at Harvey
While substantive accuracy and analytical capability is strong, qualitative evaluators noted some polish gaps relative to Opus 4.6, such as tone mismatch, a tendency to overshoot on detail, and organizational missteps such as non-standard citations.
We'll be rolling out Opus 4.7 in the model selector to eligible clients in the US and EU in the coming days. AU availability will be announced shortly.
Original source All of your release notes in one feed
Join Releasebot and get updates from GC AI and hundreds of other software products.
- Apr 16, 2026
- Date parsed from source:Apr 16, 2026
- First seen by Releasebot:Apr 17, 2026
Library on Mobile
Harvey adds a new Library tab in the iOS app, giving users a centralized, read-only place to access and reference shared prompts on mobile. The update makes it easier to reuse proven prompts and stay consistent while working on the go.
What’s New
Access and reference your shared prompts directly from a new Library tab in the iOS app.
The Library tab is now available on iOS, giving you a centralized place to access prompts you’ve shared. This initial release focuses on visibility and quick reference.
Key updates:
- New Library tab in the iOS app
- Centralized view of all shared prompts
- Read-only experience for fast access and reference
- Optimized for mobile workflows
Why It Matters
Shared prompts often power repeatable workflows across your organization. The Library tab makes them easier to access when you’re working on mobile.
With this update, you can:
- Quickly reference proven prompts without searching through chats
- Reuse existing work while drafting or analyzing on mobile
- Stay consistent with prompts already shared across your organization
This release improves access without changing how prompts are created or managed.
How to Use
- Open the iOS app
- Navigate to the Library tab
- Browse your shared prompts
- Open a prompt to view its contents
- Reference the prompt while continuing your work
Note: The Library is read-only. To edit or create prompts, use the existing prompt workflows elsewhere in the platform.
FAQ
Q: What can I do in the Library tab?
You can view and reference prompts you’ve shared. The experience is read-only.
Q: Can I edit or create prompts from the Library tab?
No. Editing and creation are not supported in the Library tab. Use other areas of the platform to manage prompts.
Q: Does the Library include anything besides prompts?
No. This initial release focuses exclusively on shared prompts.
Q: Are all my prompts available in the Library?
The Library includes prompts you’ve shared. Other prompt types are not included in this release.
Original source - Apr 16, 2026
- Date parsed from source:Apr 16, 2026
- First seen by Releasebot:Apr 17, 2026
Model Selector on Mobile
Harvey adds model selection on iOS, bringing the same flexible model controls from web to mobile. Users can switch models in-app, stay in flow across devices, and adjust behavior for drafting, analysis, or refinement without leaving their workflow.
What’s New
The model selector is now available on iOS, bringing model selection to the mobile experience.
You can now:
- Select your preferred model directly in the mobile app
- Switch models without leaving your workflow
- Use the same model selection approach available on web
- Adjust model behavior based on your task (drafting, analysis, refinement)
Why It Matters
Model selection affects how Harvey approaches your work. Bringing this control to mobile ensures you can:
- Stay in flow across devices: no need to switch to web to change models
- Match the model to the task: choose the right balance of speed, depth, or reasoning
- Work more flexibly: adjust your setup in real time as your task evolves
This update aligns mobile and web experiences, so your workflow stays consistent regardless of where you’re working.
How to Use
- Open the iOS app
- Start or open a chat
- Locate the model selector in the interface
- Select your preferred model
- Continue working — your selection applies immediately
Note: You can switch models at any time during your workflow.
FAQ
Q: Where do I find the model selector on mobile?
You’ll find it directly in the mobile interface while working in a chat or workflow.
Q: Does this work the same way as on web?
Yes. The experience is consistent across mobile and web, so switching between devices is seamless.
Q: Can I switch models at any time?
Yes. You can change models whenever your task or workflow requires it.
Q: Does changing the model affect existing outputs?
Changing the model affects future responses. Existing outputs remain unchanged.
Original source - Apr 16, 2026
- Date parsed from source:Apr 16, 2026
- First seen by Releasebot:Apr 16, 2026
March product update
Smokeball releases monthly updates that speed up email workflows, improve client intake accuracy, strengthen AML compliance tracking, and keep trust records more precise. The update also brings a new Outlook add-in, translated intake forms, and clearer form formatting.
This month’s updates help you work through emails faster, capture more accurate client information from the outset, and keeps compliance records and trust accounts in order. Keep reading to see what's new this month.
Our new Outlook add-in for Smokeball’s desktop app is here
Tier: Grow, Prosper+
Region: US, AU and UKMicrosoft is transitioning to the new Outlook, and so are we. Our latest add-in for Smokeball’s desktop app supports new Outlook for Windows and Outlook’s web app, giving you more flexibility to work wherever you need and unlocking new features such as Archie AI and tasks. Your key workflows are supported. You’ll still be able to file emails and attachments to matters, track activity with AutoTime, and send emails directly from Smokeball or Word.
If you’re using Smokeball on the Grow and Prosper+ tiers, switching now gives you immediate access to the latest features as soon as they’re released. Plus, your entire team doesn’t need to move at once; you can switch at your own pace.
We'll continue to support classic Outlook for as long as Microsoft does (at least until 2029), but no new features or enhancements will be added. Our focus is on the latest add-in, and switching sooner means fewer disruptions as Microsoft continues its transition.
If you’re on the Boost tier, you now have access to Archie AI and tasks in the Outlook add-in.
Learn more about our new Outlook add-in:
US | AU | UKSee how:
US | AU | UKAML/CTF compliance tools to help you reduce risk and admin
Tier: Grow, Prosper+
Region: AUWith Anti-Money Laundering and Counter-Terrorism Financing (AML/CTF) obligations expanding for Australian law firms, many practices are starting to think about how they’ll manage client due diligence, risk assessments, and compliance documentation.
To help, we’ve introduced new AML compliance features directly inside Smokeball. Through our enhanced integration with InfoTrack, firms can run AML checks, track risk assessments, and maintain compliance records without leaving Smokeball.
Learn more about our AML/CMF updates
See howTranslated intake forms your clients can complete with confidence
Tier: Included on Prosper+, or as part of the Intake add-on for Boost and Grow.
Region: US, AU and UKCollect more accurate, complete client information from the outset with multilingual forms. Our forms support built-in translation across ten languages, presenting questions in your client's preferred language so they can respond with confidence.
See how:
US | AU | UKYour trust records, anchored to when transactions occurred
Tier: Bill, Boost, Grow, Prosper+
Region: US and UKKeep trust records aligned with when transactions occurred, not when you entered them, by setting the transaction dates for reversals and deletions. Both the transaction date and the entered date appear in summaries and reports, giving your firm a complete audit trail. By allowing reversals and deletions to be dated to the transaction date, trust balances will be more accurately reflected at the intended point in time.
Note: Deleting a transaction reverses it and hides it; the transaction remains on record but is hidden from the default ledger view and can be made visible again at any time.
See how
Clear, formatted intake forms your clients can follow
Tier: Included on Prosper+, or as part of the Intake add-on for Boost and Grow
Region: US, AU and UKBullet points, numbered lists, and indents built directly into the form builder let you present information, and questions in a format that's easy to scan and follow. A fixed toolbar keeps those options in reach alongside existing formatting such as bold, underline, italic, heading styles, and hyperlinks, so everything you need is in one place
Original source - Apr 15, 2026
- Date parsed from source:Apr 15, 2026
- First seen by Releasebot:Apr 18, 2026
Extract File Context for Column Generation
Harvey adds file-based review table column generation, letting users quickly create or expand columns from attached files. The update supports building tables from scratch or adding columns to existing tables, with saved prompts that can be reused across different file sets.
Release Notes
Extract File Context for Column Generation
Use any file(s) as context to generate new columns in your review table
What’s New
Quickly generate review tables using your files as context to help build columns. You can ask Assistant to directly extract columns from the structure or context of files attached to your Assistant query, then build a review table from these columns or add these columns to an existing review table.
Why It Matters
This update makes it easier to quickly define columns based on your files. Whether you're building a table from scratch or adding columns to an existing one, attaching a file will consistently inform the columns Harvey creates.
How to Use
Generate Columns From Files When Building a New Review Table
From Assistant, you can use files to generate columns and build the review table at the same time. Simply add your files, send your query, and follow Harvey’s guidance to build a review table.
Files Uploaded
Example Prompt
A single file to build columns + files to be included for the table
"Extract every field specified in this lease abstract over these lease agreements"
Multiple files to build columns + files to be included for the table
"Determine whether these documents are responsive by extracting over fields specified in these discovery docs"
Only files to be included for the table (no files explicitly designated to build columns)
"Read over these files and suggest the most interesting fields to extract over and build me a review table."
Add Columns to an Existing Review Table Using files
Open an existing review table and open Ask Harvey.
Attach the files you want to use to define new columns.
Type your request and hit send — Harvey will generate new columns based on the file's contents.
Tips for Success
Save a prompt for reuse across different files.
If you’re looking to reuse your prompt for generating a review table with files as context, but the columns generated might differ depending on the files uploaded, you can save the prompt and reuse it for different sets of files.
Build your review table column prompt with a file as context.
Save it as a prompt to your Library.
The next time you need the same structure, pull up the saved prompt and attach a different set of files — Harvey will generate columns based on whatever files you provide.
FAQs
Q: What file sources can I use to build columns?
The column file source must be the same source as the files for your review table. For example, you will not be able to generate columns if you add a template that’s uploaded directly from your computer and files attached from Vault; you would need to upload that template into the same Vault to be able to extract columns from the template alongside the Vault files used in the table.
Original source - Apr 15, 2026
- Date parsed from source:Apr 15, 2026
- First seen by Releasebot:Apr 17, 2026
Edit & In-Document Citations
Harvey for Word introduces expanded citation support, linking answers and suggested edits to source material and the active Word document. New in-document and edit citations improve transparency, speed up verification, and help users review revisions with confidence.
Harvey for Word will introduce expanded citations capabilities, expanding transparency by linking answers and edits directly to source material and the active Word document.
What’s New
Harvey for Word has expanded citation coverage to provide greater transparency with two new citations types:
- In-Document Citations link answers directly to relevant sections of the open Word document
- Edit Citations attribute suggested changes to uploaded documents or precedent
Together, these enhancements will make it faster to verify answers, review edits, and maintain confidence in document accuracy—without interrupting drafting workflows.
Why It Matters
Expanded citation coverage will improve transparency, making it easier to validate outputs without manually searching for supporting language. Trace answers back to specific paragraphs or references and view exactly which source material informed a suggested revision.
How to Use
Once you’ve run a query, you can click any numbered citation to jump directly to the reference in your document.
If you receive suggested revisions based on uploaded precedents or documents, those suggestions will contain numbered citations as well.
FAQs
Q: How are In-Document Citations different from existing citations?
Existing citations link to uploaded documents. In-Document Citations will additionally link answers to specific sections of the active Word document, allowing you to verify responses against the document they are currently editing.
Q: Who will have access to these new citation types?
These citation enhancements will be available to all Harvey for Word users.
Original source - Apr 15, 2026
- Date parsed from source:Apr 15, 2026
- First seen by Releasebot:Apr 16, 2026
From First Pass to Final Sign-Off: Review Tables That Keep Legal Teams in Sync
Harvey adds smarter, faster review tables for large-scale document review, with real-time collaboration, improved extraction logic, conditional columns, sentence-level citations, and easier table setup to help legal teams surface insights and align on decisions.
Harvey's latest review table improvements bring real-time collaboration, smarter extraction logic, and faster table setup to large-scale document review.
Legal teams running large-scale diligence, contract portfolio analysis, or litigation document review need more than just raw extraction power. They need a collaborative source of truth to quickly surface findings, stay aligned as a team, and sign-off on key decisions in one place
Harvey’s review tables are now easier to set up, more intelligent, and more interactive, helping legal teams sift through vast data sets to surface the insights that matter.
Faster Review Setup
Imagine you’re an associate kicking off M&A due diligence with a data room of thousands of documents. You upload your files to the review table and start building a table by simply typing “change-of-control provisions.” To refine your request into a more precise and effective prompt, the Improve Prompt feature rewrites your query to extract the terms you need.
To accelerate the rest of the table build, you upload an existing diligence request list and Harvey automatically translates the fields of the template into columns. Instead of building from scratch, the table now reflects how your team already runs diligence.
Surface Richer Insights From Your Documents
As documents are processed, the table doesn’t just fill with static extractions. With conditional columns, you can extract governing law in one column, then trigger jurisdiction-specific checks in the next. Or you can identify whether a contract includes an assignment clause, and extract consent requirements only when it contains that clause. Each column builds on the other, so the table reflects how a diligence associate actually thinks: not as isolated questions, but as a chain of reasoning.
At the cell-level, an improved review algorithm reduces time to complete review with model reasoning displayed and sentence-level citations.
Align Your Team Inside the Review
Once the table is fully populated, you loop your team in for real-time collaboration. Use multi-colored flags to triage issues: red for deal-breakers, orange for negotiation points, and yellow for standard terms, creating a visual map of risk across the dataset. With comments, you can explain why something was flagged, capture nuance like “assignment requires counterparty consent,” or resolve open questions with your colleagues. You can also use manual input columns to add commentary for each row directly onto the table for you to keep track.
At any time in the review, you can ask questions like “summarize all red-flagged tenant obligations” and get answers grounded not just in the table, but in metadata that capture your deal team’s thinking, including flags, comments, and verification.
Instead of splitting work across spreadsheets and emails, review tables function as a single source of truth for large-scale analysis and collaborative review, driving faster review cycles and higher-confidence outputs.
If you’re interested in learning more about how Harvey can help your team, reach out to us below.
Original source - Apr 15, 2026
- Date parsed from source:Apr 15, 2026
- First seen by Releasebot:Apr 16, 2026
Admin Analytics Dashboard Improvements
Harvey introduces a redesigned Analytics dashboard with richer usage insights, trend views, and executive-ready reporting for admins and Knowledge Managers. The update adds clearer engagement tracking, active user trends, and more reliable reporting across dashboard, export, and API.
What’s New
We’re introducing a redesigned Analytics dashboard with richer usage insights, trend views, and executive-ready reporting. These updates will help organizations better understand engagement, demonstrate ROI, and proactively drive adoption.
Harvey's Analytics dashboard has been redesigned to give admins and Knowledge Managers clearer visibility into usage and adoption. Flexible time-based views and filters make it easier to track engagement trends, identify top users, spot adoption gaps early, and pull together renewal-ready reports.
Updates include:
- Granular, event-based tracking — Captures activity at the event level across product surfaces and subsurfaces, with identifiers for workflow agents, playbooks, vaults, and review tables.
- Clearer activity distinctions — Explicitly separates new vs. follow-up activity, and distinguishes Create vs. Run actions for workflow agents and playbooks.
- Consistent reporting schema — Unified data structure across the dashboard, export, and API so metrics are reliable and comparable however you access them.
- Active user trends — DAU, WAU, and MAU views with rolling time-period selection and expanded filtering for flexible analysis.
Why It Matters
As an admin or Knowledge Manager, you now have a clearer view of how your team is using Harvey. Now you can better track engagement trends, demonstrate ROI to stakeholders, and spot under-engaged users early.
How to Use
Refer to View Analytics and Reporting for steps and tips on navigating the dashboard.
FAQs
Q: Who is this dashboard designed for?
The dashboard is built for Admins and Knowledge Managers who need a self-serve view of team engagement. It provides a centralized way to monitor adoption and understand how Harvey is being used across the organization.Q: How will this help demonstrate ROI?
Original source
By surfacing clear usage trends, active user metrics, and high-impact workflows, the dashboard will make it easier to quantify engagement and highlight business value in executive and renewal conversations. - Apr 15, 2026
- Date parsed from source:Apr 15, 2026
- First seen by Releasebot:Apr 16, 2026
Guest Accounts
Harvey adds Guest Accounts for secure collaboration in shared Spaces, letting organizations invite external partners or internal cross-functional users with limited access while keeping security, privacy, and data ownership in place.
Guest Accounts let your organization securely collaborate with external partners — such as clients, outside counsel, or advisors — inside a shared Harvey Space, without requiring them to become full Harvey customers.
What’s New
A Guest Account is a limited access account that allows organizations to invite external collaborators — or, for in-house teams, internal cross-functional users — into a Shared Space. With Guest Accounts, you can:
- Collaborate with external partners (non-Harvey customers) directly in Harvey at no additional cost
- Limit access only to the Spaces, tools, and artifacts you intend to expose
- Maintain security, privacy, and data ownership within your organization
Why It Matters
Guest Accounts enable collaboration without requiring both parties to be Harvey customers, unlocking workflows like contract review, policy Q&A, and intake.
How to Use
Refer to our new article, Guest Accounts Overview, for full steps, FAQs, and tips. You can also learn more about Security, Data, and Privacy for Sharing in Harvey here.
Original source