New updates and improvements about V7 Go

You can now favorite any workflow or chat and get back to it in one click. Hit the star on a card to add it, and it appears in the Favorites section of the left navigation alongside everything else you have starred.

What they are useful for

• Recurring work: star the workflows you run on a cadence, like a weekly reconciliation or a monthly report, so you are not searching for the same one every time.

• Picking up where you left off: star a chat and return to that exact session later. The context is already there, so a similar task continues in the same thread instead of starting cold.

• Bookmarking a working setup: when a workflow is configured the way you want it, star it. It becomes the version you come back to and adapt, rather than one you rebuild.

Favorites are per user, so starring something does not change what your teammates see in their own navigation. Add and remove them as often as you like.

Three releases today, and the theme is seeing what changed and being able to undo it. Workflow properties keep a version history now, and there is an inspector to compare versions and preview a restore before you commit to one.

Versions and history

• Workflow property versions: Workflow properties keep a version history. You can see how a property changed over time instead of guessing which edit broke a run.

• Version inspector: Compare two versions side by side and preview what a restore will do before you apply it. Labels in the navigation rail are clearer, so you can tell which version you are looking at.

Reporting and calculations

• Hourly usage reporting: Usage data breaks down by the hour, and transformation progress appears in reporting. An activity spike is now something you can locate in time.

• Calculations that recover: A calculation that hits a database error resumes instead of failing. Long-running work continues rather than restarting from the top.

Stability

• Structured search stability: Searches that use JSON schema validation complete instead of crashing. You can keep narrowing a search without losing your place.

• Property changes apply immediately: A change to a property's JSON configuration is detected properly. No more cached configuration quietly overriding what you just set.

Interface

• Overview columns stay put: Your column choices in Overview survive a page reload. Set the table up once and come back to the same view.

• Refreshed date picker: The date picker matches the current design across the product. Scanning a calendar and choosing a date is a cleaner job than it was.

• Custom MCP connector badges: Status badges in the custom MCP connectors table size correctly, so the table scans evenly.

The last stretch of July leans towards the plumbing that keeps V7 Go connected to everything else. MCP connections can authenticate with an API key, images and PDFs now go through the same vision path, and nothing a run produces gets lost on the way out.

MCP and integrations

• API key authentication for MCP: MCP connections can authenticate with an API key. This helps where OAuth is impractical, such as a service account or a headless deployment.

• MCP OAuth discovery to spec: OAuth metadata is discovered in the order the MCP specification requires, so compliant services connect on the first attempt.

Documents and files

• One vision path for images and PDFs: Image crops and PDF pages run through the same vision processing. Results stay consistent across both, so you do not need to prepare a scan differently from a photo.

• Automatic case files: Files a run creates are saved as case files when the run finishes. Nothing has to be rescued by hand from a completed run.

• Artifact links become file cards: An artifact path containing a space or ending in punctuation renders as a file card instead of plain text.

• Clear out stuck library files: Delete library files that got stuck while processing. A blocked upload is now something you can clear yourself rather than a support ticket.

Budgets and reports

• Budgets in Mission Control: View and manage budgets from Mission Control, next to the work the spend belongs to. Budgets themselves shipped on 14 July; this is where they now live.

• Reports across workspaces: Mission Control reporting runs on the multi-workspace API, so a report can cover more than the workspace you happen to be in.

• Longer report date ranges: Mission Control reports accept extended date ranges. A quarter of activity fits in one report instead of three stitched together.

Sessions

• Cleaner session expiry: An expired session now ends in a consistent sign-in prompt. You get a way back in rather than an error screen from whichever feature noticed first.

Two new Gemini models, a better answer to "where did our tokens go", and one fewer reason to email support: organization admins can change their own settings now.

Models

• Gemini 3.6 Flash and Gemini 3.5 Flash Lite: Both are available for extraction and AI work. Flash Lite is the cheaper choice when the task is simple and the volume is high.

Cost and usage

• Token usage by dimension: Group token usage by the dimensions your team cares about instead of reading a single workspace total.

• Cost attribution for extraction and calculation: Usage detail shows which work created which cost, so an expensive extraction is traceable to the thing that ran it.

• Usage stats in Mission Control: Activity and key numbers appear in the Mission Control overview, rather than on a separate page.

• Reports that reflect reality: Reports show live deletion status and budget detail in the results, so you are not cross-checking against something else.

Administration

• Organization settings, self-serve: Organization admins manage their own settings without a support request. Workspace mappings and reporting access stay consistent across regions.

• Property history and restore: View a property's previous configurations and restore one when a change turns out to be a mistake.

• Email replies per workspace: Set how email replies behave for each workspace instead of applying one rule everywhere.

Fixes

• Citation highlights on scans: Citation highlights line up correctly on scanned PDFs and image-only documents, where there is no text layer to anchor to.

• Delete property dialog: The dialog shows its full warning and buttons. You can read what you are about to delete before you confirm it.

This stretch is mostly reporting and cost control. Reports got a dedicated overview, admins got finer control over who sees them, and runs now stop themselves before they get expensive.

Reporting

• Reports Overview: A dedicated section for reporting, so there is a place to start rather than a set of separate pages to hunt through.

• One report across workspaces: Generate a single consolidated report covering several workspaces in the same region, instead of pulling exports and joining them by hand.

• Sortable workflow reports: Sort, filter, and page through workflow detail reports. This matters once a report runs to thousands of rows.

• Report access at organization level: Admins can grant reporting visibility to specific people rather than opening it to everyone.

Cost and limits

• Run limits: A run stops before it runs too long or consumes too much. A misconfigured loop now costs you minutes.

• Chats stop at the billing limit: Once a workspace reaches its billing limit, chats stop. The limit is now actually a limit.

• Calculations that finish under load: Calculated fields complete when many updates land at once, instead of sticking half-done.

Reading and review

• PDF citations out of the way: Citation markers on a PDF no longer cover the text underneath.

• One source's markers at a time: Select a source and only that source's citation markers stay visible. Everything else steps back.

• Code blocks left alone: A code block in chat stays exactly as written. No more "Unknown item" pills appearing inside a snippet.

Five releases over the weekend and Monday. The one worth reading twice: a case can now stop and ask you a question mid-run rather than guessing at the answer.

Models

• GPT-5.6: Pick GPT-5.6 from the model picker when you set up a chat or a task. Prompt caching works correctly with it, so responses come back without the loading stalls the first builds had.

Cases and workflows

• Questions inside cases: A case can pause and ask you for input before it continues. Required questions stay in front of you until they are answered, so a run does not proceed on a guess.

• Field recovery: A project with recalculation switched off can recover fields that never finished computing. No manual repair, no blank column.

• Folder sync controls: Turn sync on or off for one folder at a time instead of all or nothing.

• Required integrations, connected in place: When a workflow needs an integration you have not connected yet, you can connect it without leaving what you were doing.

Evidence and citations

• More accurate citations: Citations match the source text more precisely, including text inside web pages. The stray anchor text, extra blank lines, and misplaced highlights are gone.

• Citations in generated artifacts: A generated page carries its citations in the content, so you can check a figure without reopening the source pile.

Reporting and scripts

• Favorites sidebar: Favorites live in the sidebar, persist across sessions, and can be added from context menus. There is a documented API for them too.

• Python scripts that take input: A Python run can accept input while it runs, so scripts that prompt for a value or read pasted data work as written.

• Workspace usage reports: Token usage reports name the workspace on every row. Admins stop stitching per-workspace exports together.

Two backend releases to start the week. The headline is a native Google Drive picker, so documents you already keep in Drive no longer need a round trip through your downloads folder. Extraction got faster, and every run now reports what it cost.

Getting documents in

• Google Drive picker: Pick files straight from Drive with a native picker. No downloading and re-uploading to get a document into Go.

• Faster extraction: Documents process faster during extraction. The difference shows most on large batches.

• Extraction that matches your structure: Extraction results follow the data structure you defined more closely, which makes outputs easier to compare and review.

Models and runs

• Claude Sonnet 5: Available in the model picker for chats and ad-hoc tasks where you want stronger reasoning.

• Token and cost tracking per run: Every run reports its token usage and cost. You can see which work is expensive before the invoice tells you.

• Warnings when an answer is cut short: If a response hits the output limit, you are told. A truncated answer used to look like a finished one.

• Long runs that finish: Runs no longer crash when a model reaches its token limit mid-action. Long tasks fail cleanly instead of dropping everything.

Review and status

• Live queue status: Queue state updates as work starts moving, with no page refresh.

• Cleaner evidence highlights: Source highlights no longer stack duplicate boxes around the same passage.

• Python results render: Output from a Python run displays correctly when the run returns files.

Budgets are now available to every workspace. Set spend limits on AI activity, watch consumption in real time, and stop runaway costs before they happen — all from Settings → Budgets.

Two levels of control

Workspace budgets cap total spend across all Go activity. Agent budgets set per-agent limits scoped to a specific workflow, each with its own live tracking — so one heavy agent can have a tighter leash than the rest.

Dollars or tokens

Every limit and progress bar can be viewed in dollars or in tokens. Dollars is usually the easier number to reason about — you set a cap of $5,000, not a count of tokens — but the token view is there when model consumption is what you care about.

Periods and overrides

Choose a reset cadence of daily, weekly, or monthly. Each period must be unique per policy, but you can stack several on the same agent — a daily and a weekly cap together, for layered protection. A default limit refreshes automatically each period; a current-period override adjusts spend just once, without changing the standing default.

Tracking and alerts

A live progress and a widget in each agent view show current spend, the limit, and the reset date at a glance. Email alerts fire automatically at 75%, 90%, and 100% of a budget, and full spend reporting lives in Reports.

Reports now go down to the level of a single chat. Alongside the existing agent and user views, you can see exactly what each conversation cost — in tokens and in dollars.

For any chat, you get its token spend and dollar value, a breakdown of which models were used, and whether any Go agents were triggered during the conversation. An expensive chat is no longer buried inside a workspace total — you can see which conversations are driving cost, and why.

Open Reports in the left navigation and select the Chats view. The functionality is available to workspace administrators.

Agent memory lets you teach V7 Go how you like to work. Write your standing instructions once and Go carries them into every chat, so you stop re-explaining the same preferences at the top of each conversation.

It works like a set of house rules that sit behind your prompts. Anything that is true for most of your work belongs here: the register you write in, the way you want answers structured, the checks Go should run before it commits to an output, or the details it should always ask for.

Keep it short and specific. A handful of clear rules works better than a long wish list. Open Settings → Agent memory, write your guidance, and save.

V7 Go is now available as an MCP server. Connect it to Claude, ChatGPT, or any MCP-compatible assistant, and you can run Go straight from the chat window without switching back to the Go UI. This is the reverse of connecting tools into Go: here your assistant becomes a front end to Go itself.

What you can do from chat

• Inspect and query workflows: pull rows and entities, run queries over workflow data, and read definitions to answer questions like “what did this run output?”

• Run workflows: kick off Go runs from chat, including ones that take file inputs.

• Build and edit workflows: add and update properties, including computed and agentic ones, configure views with filters and sorts, and set up triggers, all conversationally.

• Query a Context Graph: ingest documents into a Context Graph and ask questions in natural language.

Connecting

You authenticate with your V7 Go login over OAuth when you connect. Each assistant configures MCP servers a little differently. In Claude, open Settings → Connectors → Add custom connector.

Endpoints

Pick the one that matches your account’s region:

• EU / standard: https://mcp.go.v7labs.com

• US: https://mcp.go.us.v7labs.com

You can now open a chat window directly against any Hub and ask it anything — from a quick question to a prompt that spans every document in it.

When a Hub is created, V7 Go indexes every file in it. By the time you open the chat, the agent already understands what's there and can select the most relevant files for your specific question immediately. Send a complex prompt and you see the reasoning unfold in real time — which files it's drawing from, what it's extracting, and where it is in the task.

Chatting with Hubs is built primarily for the ad hoc questions that arise between defined workflows. When your team has a repeatable process, agentic workflows handle it. When an analyst needs a one-off deep-dive on a specific company, or a question comes up mid-review, this is what they reach for. Hubs are the first surface — expect the same contextual assistant to appear in other parts of the platform soon.

To access it, open any Hub and look for the chat panel on the right.

Usage reporting now includes a live dashboard — so you get a clear picture of how AI is being used across your workspace without exporting a CSV first.

The Reports section has two views. The Agents tab shows which workflows are running, which models they use, how many tokens each consumes, and what each costs over any date range. If one agent is doing the heavy lifting and three others are idle, that's visible the moment you open the page.

The Users tab shows individual engagement: fields edited, chats started, messages sent, corrections made, and last activity. It tells you not just whether people are logged in, but whether they're getting value from the platform.

CSV export in both Reports and Billing and Usage remains available for when you need it — bulk analysis, sharing with finance, or building your own views. The dashboard is for the daily question: how is AI actually being used, and by whom.

Go to Reports in the left navigation. Both tabs are available to workspace administrators by default.

V7 Go now supports MCP (Model Context Protocol) integrations. Your chats and agents can pull live context and functionality directly from the tools your team already uses.

The initial set includes HubSpot, Notion, Attio, Granola, Linear, and Slack. Once connected, the agent can search your CRM, pull meeting notes, create issues, query documents, and more.

Adding custom MCP servers

Beyond the built-in connectors, you can bring your own. Go to Integrations → MCP Connectors → + New to register any MCP-compatible server by URL. Once added, it appears in the Custom section and can be toggled on per conversation like any other connector.

For additional setup instructions, see the MCP Connectors documentation.

V7 Go can now generate fully formula-linked, multi-sheet spreadsheets. You can preview them as an interactive artifact and download the output as an Excel file. To use it, type / when writing your prompt and select the Spreadsheet Generator from the menu. From there, describe what you need and the agent handles the build.

The Spreadsheet skill is purpose-built for structured financial and analytical work. Give an agent a brief and it will produce a complete workbook: cover sheet, assumptions, income statement, DCF model, sensitivity analysis, comparables, and more. This functionality can be even more powerful when combined with specialized AI agents.

This is one of many skills available in V7 Go, with new ones added regularly. Press / in the homepage prompt box at any time to browse the full list.

You can now edit slides directly in V7 Go.

Our slide generation skill already makes it possible to create complex presentations with charts, structured content, and polished layouts. Combined with AI agents for workflows like financial due diligence, it can generate things like investment decks in minutes.

But generation is only part of the workflow. Sometimes you want to fine-tune the output yourself.

Now you can choose between updating slides with prompts or editing them manually. Just switch from View to Edit mode to change text, update styling, and refine slides directly in the editor.

When you’re done, you can also export your presentation as a downloadable PPTX file.