External APIs for Enterprise Agents - connect third-party API credentials to your agent without ever pasting a key into chat. A major security upgrade for agents that reach outside services.
Store credentials in a secure vault, encrypted at rest - the same level of protection as environment variables on a platform like Vercel.
The agent only ever receives a short-lived proxy key. It never sees your real key and cannot leak it to end users, even if asked directly. The security comes from the proxy, not from prompt instructions.
Pick from a curated list of platforms and just add your key, or bring any OpenAPI schema via URL, file upload, or paste. The agent figures out how to use it.
Lives on a new tab on the Actions page. Requires the Smart Tasks action with external permissions enabled. Available on all plans.
🛠 Improvements
The agent chat interface now fully localizes to the agent's configured language. Recently added features - file manager, conversation history, and the rename/share/export actions - were still showing in English and are now translated.
Long chat responses now stream faster all the way to the end. Previously, rendering slowed down as the response grew longer.
You can now link directly to a specific Personalize tab (Persona, Citations, Advanced, and so on) through the page URL, so a tab can be bookmarked or shared and opens instantly.
Citation highlighting in the PDF viewer is more accurate, including when the highlighted passage runs across two pages.
Export your Outgoing Traffic (Links & Sources) analytics to CSV. The export runs in the background and emails you a download link, matching the filters and date range on screen.
Files generated by an Enterprise Agent inside a Slack conversation now arrive as native file attachments in the thread.
Enterprise Agents task plans now generate more complete to-do steps, and the agent more reliably picks the right action (web search, code execution, or deep search) for a given question.
🔌 API & MCP
Added GET and PATCH /api/v1/projects/{projectId}/deployment to read and update an agent's embed, live-chat, AI-assistant, instant-viewer, and SGE widget configuration. PATCH updates one section at a time via the {section} path parameter.
Added read-only billing endpoints under /api/v1/billing/ for current plan, payment methods, invoice list, and invoice download.
The Customer Intelligence reports endpoint /api/v1/projects/{projectId}/reports/intelligence now accepts a human_request filter and returns human_request on each result.
🐞 Bug Fixes
Fixed the issue where teams on a plan with add-ons could not resume a cancelled subscription.
Fixed the issue where a team's plan limits and pricing could stay outdated after a custom plan was attached or removed.
Fixed the issue where Enterprise Agents code execution failed on every Python step.
Fixed the issue where an Enterprise Agent could not read a file it had just generated earlier in the same task.
Fixed the issue where Enterprise Agents responses showed escaped characters instead of formatted text.
Fixed the issue where larger Enterprise Agents dashboards could render broken when split across separate files.
Fixed the issue where the status message shown before an Enterprise Agents step sometimes described a different action than the one being run.
Fixed the issue where a citation image that failed to load briefly showed an empty gap instead of a fallback icon.
Fixed the issue where the task status spinner in Enterprise Agents appeared clipped instead of a full circle.
New AI models: Claude Opus 4.8 and Gemini 3.5 Flash are now available for your agents.
Claude Opus 4.8 is available to everyone with model selector access, at the same cost as older Opus models (1 extra query credit). Opus 4.6 is now tagged legacy.
Gemini 3.5 Flash replaces Gemini 2.5 Flash, which is being retired by Google.
Models retired by providers - Claude 3.5 Haiku, Claude 4 Sonnet, and Gemini 2.5 Flash - have been removed, and agents using them were automatically moved to the newest model from the same family. No action needed.
🛠 Improvements
AI Disclaimer: add a disclaimer (up to 160 characters) to your agent, shown before the first message in every conversation. Configure it under Personalize > Advanced. Available on Premium.
Auto-Sync add-on: Standard and Premium teams can now unlock automatic re-syncing for Google Drive, OneDrive, SharePoint Docs, SharePoint Sites, Confluence, Notion, and Zendesk sources. Available through the Contact Sales flow on the Billing page.
Download dashboards: save any dashboard generated by Enterprise Agents as a single self-contained HTML file that works offline - including dashboards with separate data files and non-English filenames.
The Slack integration now works with Enterprise Agents - follow-up replies in a Slack thread continue the same conversation, so the whole thread appears as a single session in your dashboard.
CSV files generated in chat now display as a proper table with horizontal scrolling, instead of raw comma-separated text.
Long chat responses now stream smoothly to the end - previously, streaming got slower as the answer went on and could keep "typing" long after the response was done.
Custom Action (MCP) connection tokens now refresh automatically - no more warning asking you to reconnect when a token expires.
The Custom Action popup no longer mentions the 3-use limit for Enterprise Agents, where the limit doesn't apply.
🔌 API & MCP
POST /api/v1/projects/{projectId}/settings now returns 207 Multi-Status when some submitted fields were ignored, with the response listing updated and ignored fields. Requests where all fields are valid still return 200.
User query limits are now enforced on the Tasks API - POST /api/v1/projects/{projectId}/tasks and POST /api/v1/projects/{projectId}/tasks/{taskId}/messages return 429 when the user is over their usage limit.
🐞 Bug Fixes
Fixed the issue where setting a daily query limit for the Guest role also rate-limited team members with other roles, such as Team Administrators.
Fixed the issue where certain web pages - such as those behind redirects or requiring full page resources - weren't fully captured during indexing, leaving the agent unable to answer questions about them.
Fixed the issue where Google Drive appeared locked when added from the Documents tab, even though it worked from the Sources tab and Create page.
Fixed the issue where Plan & Act Mode incorrectly appeared as an option in the Add Source menu.
Fixed the tooltip on the Custom Action usage notice being too narrow for its text.
Fixed awkward mid-word line breaks on the Custom Button action card.
Zendesk integration, rebuilt on OAuth. Connecting Zendesk now uses a one-click OAuth flow instead of manual API keys, and the app is officially approved in the Zendesk marketplace.
Connect Zendesk Tickets as a knowledge source, alongside Knowledge Base articles.
Ticket anonymization strips customer emails, names, and IDs from processed tickets, so you can train an agent on tickets and deploy it publicly without exposing user data.
The previous API-key integration is now labeled Zendesk (Legacy) and hidden from the Create page and Add Source menu. Existing Legacy connections keep working with no action required, though migrating to the new OAuth version is recommended.
🛠 Improvements
The Personalize page header now shows just the agent name.
Switching teams now takes you straight to the Dashboard, instead of leaving you on the Teams page with no feedback.
The Custom Action (MCP) setup modal has been polished: field placeholders, right-sized warning text, the Description section open by default, and a note that your token is stored encrypted and never shown to end users.
🐞 Bug Fixes
Fixed the issue where saving persona instructions could intermittently fail with a 500 error when the agent was updated by two requests at the same time.
Fixed the issue where searching on the Agents page returned an empty list if you were already on page 2 or later - search now jumps back to the first page of results.
Fixed the issue where duplicating an agent carried over incorrect field values, such as source indexing timestamps that predated the new copy.
Enterprise Agents are now live on every plan. Plan & Act mode lets an agent plan a multi-step task, run actions and tools, and stream results back as it works. Available everywhere, with a per-query task ceiling of 5 on Standard, 10 on Premium, and 15 on Enterprise.
Plan & Act Mode card on the Create page spins up an empty, action-first Enterprise Agent without a knowledge base.
Empty-knowledge-base agents now work on deployments, matching in-app behavior.
Action limits behave gracefully - when the agent hits its per-query action budget mid-task, it stops cleanly, returns partial findings with citations, and asks if it should continue. Resume picks up the cached plan.
Persona-declared source priority is now respected during retrieval, so the planner checks the files your persona names first.
Reconnect banner appears with a one-click Reconnect when the streaming connection drops on a long run.
Free label replaces 0 for zero-cost RAG tools in the Plan & Act breakdown.
Default toolset enables from the start, regardless of plan, and self-service plans no longer cap the number of actions an Enterprise Agent can enable.
Stricter task labels - the agent no longer claims to have built a dashboard or report when no file was generated; non-English users now get task labels in the right language.
Enterprise Agents tag marks these agents on the Agents page, alongside Multi-Agent.
OAuth for Custom Actions (MCPs). When adding a Custom Action, you can now pick the auth type: None, Token, or OAuth. This makes the most popular MCPs - Notion, Zapier, and the like - trivial to connect, no manual token handling. Agent behavior itself is unchanged; this is purely an auth upgrade.
Export Conversation Leads (Zapier). New Zapier action exports the merged lead from a single conversation by session ID, with an optional row-by-row mode.
🛠 Improvements
End-user login via identity provider: fewer clicks to sign in, plus an optional JWT auto-login flow for customers who authenticate users on their own backend and pass a signed JWT into the embedded agent.
Agent search on the Agents page now matches both agent name and agent ID - paste a partial ID and the matching agent appears.
Google Drive: when authorization expires (revoked permissions, missing refresh token, etc.), users are now notified in-app and via email with a one-click reauthorize popup, matching the other source integrations.
The pagination component has been rebuilt with a consistent UI across the app: highlighted current page, capped visible page count, ellipses for skipped pages, and Previous/Next controls.
Document Analyst settings now expose a single "Allowed document length" field that applies to both knowledge base files and uploaded attachments, replacing the two separate length controls.
The redundant tab row is hidden on the Multi-Agent Build page (only one tab there).
YouTube source: timestamps are tighter and indexed sections now include richer descriptions, giving search better hooks into video content.
🔌 API & MCP
New endpoints under /api/v1/projects/{projectId}/children to list, add, remove, and bulk-unassign child agents on a Multi-Agent, with the same gating and limits as the web UI.
New endpoints under /api/v1/projects/{projectId}/contextual-rag to list, delete, and bulk-update Webpage Awareness page messages.
GET and POST /api/v1/projects/{projectId}/settings reached parity with the web UI - 9 previously web-only fields are now readable and writable, including max_actions_per_query, is_pdf_viewer, conversation_history, and allowed_domains_for_embedding.
Deprecated Agent Plugins (ChatGPT plugin) endpoints and schemas have been removed from the public API reference.
MCP server upgraded to mcp 1.27.1 - DNS rebinding protection is now enabled by default.
🐞 Bug Fixes
Fixed the issue where chat file uploads silently failed on certain valid file types: .xlsx and .csv attachments are now accepted, markdown files containing HTML-like tags (iframes, scripts) are no longer rejected, and CSV files served as application/csv instead of text/csv no longer return a 500. .html is also now accepted by Smart Tasks Read Files.
Fixed the issue where the Task API silently dropped files[] uploads instead of attaching them to the request.
Fixed the issue where HTML entities like ', &, and " rendered as literal text in the User Query and AI Response columns of Customer Intelligence (common with non-English content).
Fixed the issue where guests couldn't sign in with Google on private shared agents served from *.customgpt-sites.com subdomains.
Fixed the issue where clicking an already-used email verification link rendered a broken page showing both "Email verified" and "Invalid user ID" - users are now redirected cleanly to billing, dashboard, or login with a clear message.
Fixed the issue where an end-user signing in via identity provider through /portal could hit a 403 when opening a shared agent (the embedded flow worked).
Fixed the issue where the Multi-Agent stats card always showed 0 query credits and per-child counts ignored agentic query multipliers. Labels now read "Query Credits (This Billing Cycle)".
Fixed the issue where a Multi-Agent with no child agents showed an empty "Choose the agent" modal and a stuck loading spinner in the dropdown.
Fixed the issue where a broken citation image URL surfaced as a broken thumbnail in answers - failed images are now hidden per citation.
Fixed the issue where filter dropdowns on the Agents page went empty after deleting or replicating an agent.
NextGen agentic task failure states now render in amber instead of red, signaling that a partial failure is not critical.
🔌 API & MCP
POST /projects accepts a new optional use_nextgen_agentic_flow boolean. Set it to true to create an agent with the NextGen agentic flow enabled from the start, instead of toggling it separately. Defaults to false.
🐞 Bug Fixes
Fixed the issue where pagination in the Customer Intelligence explorer returned empty results on page 2 and beyond, with incorrect total-page counts.
Fixed the issue where query credits in the analytics dashboard were undercounted, sometimes showing a credit total lower than the actual number of queries run.
End-user usage limits - team admins can now cap how many queries individual end users consume, in query credits, on a daily and monthly basis.
Limits can be set per user role, per individual user, or for guest visitors. Individual limits override role limits when both are set.
The block message shown when a user hits a limit is fully customisable.
Guests are identified by browser session for now - not bulletproof, and customers should be aware.
Available to teams on the CustomGPT.ai for Teams Enterprise plan feature.
🤖 NextGen Updates
Working Documents UI: pinned to-do list - the agent's live to-do list is now pinned above other working documents, and any [TODO.md] mention in the agent's reply becomes a clickable chip that opens the same file. The list updates in real time as the agent plans and acts.
When NextGen runs an expensive reasoning model inside a Smart Task, the added query credit cost for the model now a 3x multiplier.
Plan preview + contemplation step - NextGen now shares a brief plan preview at the start of complex tasks, and contemplates before acting on multi-step work. This gives end users earlier feedback on what the agent is about to do.
Document Analyst accuracy improvements - the Document Analyst step that pulls and reasons over knowledge-base files has been tightened to reduce hallucinations when the question spans context, file analysis, and web search at once.
Document Analyst configuration polished - the source-picker layout and copy have been improved.
🛠 Improvements
Queries vs. query credits - the in-app pricing language has been clarified. What used to be called "queries" everywhere is now called "query credits" when it refers to the cost charged against your plan. The Dashboard, Analyze, and Agents pages now show both numbers side by side when they differ - so you can see how many actual queries your users sent, and how many credits those queries consumed. Nothing about plan limits or billing has changed.
Customer Intelligence: Requested Human metric - a new Requested Human column and filter has been added to Content Found, User Emotion, and User Intent views, plus the conversation insights modal. If an end user asked to talk to a human (in any wording), this metric flags it as Yes.
New agent source: Website (Low Performance) - a new agent source for websites whose servers can't handle the regular crawler. Requests are spaced out so pages get indexed reliably; ingestion is slower but completes.
Source label limits expanded are now expanded up to 50 labels per document and label names up to 100 characters.
Websites that block our sitemap probe now fall back to a scraping path, so their pages get indexed instead of being silently dropped.
Tooltip width and max-height of Inline citations popup were tuned so long previews stay scannable.
Custom Action name and description length validated inline - name (50 chars) and description (255 chars) limits are now surfaced in the modal before save.
The in-app copilot's send button is now disabled on whitespace input.
🔌 API & MCP
Full API support for agentic actions configuration - the API now lets you do everything the UI does for agentic actions:
Toggle any agentic action on or off.
Configure each action's settings.
Add a Custom Action (MCP) directly through the API.
Same rules as the UI are respected - compatible actions per agent version, plan-level concurrency limits, etc.
GET /projects/{projectId}/actions returns a slim payload with key, name, is_enabled, and settings. Sensitive tokens come back obfuscated as TUOR…1Zdw.
Task endpoints now support AND/OR label filters - POST /tasks and POST /tasks/{taskId}/messages accept the same grouped labels[] AND-of-ORs format as /conversation/messages, instead of OR-only.
Query credits enforced on Task API - POST /api/v1/projects/{id}/tasks and /tasks/{taskId}/messages now check remaining query credits before dispatching work, matching the behaviour of the conversation endpoint.
GET /reports/leads - filter by conversation - a new optional session_id query parameter limits the export to leads captured in a single chat session.
Leads export OpenAPI fixes - the response now correctly includes user_defined, and both user_defined and tags are documented in the OpenAPI schema.
🐞 Bug Fixes
Fixed the issue where the NextGen agent could break mid-stream on responses with many citations, causing the answer to fail to complete on the user's screen.
Fixed the issue where the NextGen Documents Analyzed stat read 0 even when the Document Analyst was clearly used. The count is now correct.
Fixed the issue where the agent's to-do list could stop updating mid-stream after its first change in a NextGen conversation.
Fixed the issue with NextGen Configuration Summary on the Intelligence tab.
Clicking the Verify Responses badge on a NextGen reply now lands on a correct error page.
Fixed the issue where the "Actions & tasks" label was missing from the in-app agent preview panel in Chatbot Settings and Tools & Actions. It now matches the live Ask page.
Fixed the issue where inline citation rendering was broken in some edge cases on the NextGen response stream.
Fixed the issue where adding a new source to an existing agent wouldn't refresh the sources list header until reload.
Fixed the issue where premium-only project settings stayed enabled after a downgrade from Premium to Standard. Settings are now consistently normalized to the Standard defaults on downgrade.
A Multi-Agent with no child agents now displays a clear empty state with a path to add child agents, instead of erroring out on send or when unassigning the last child.
Failure feedback in the agent UI - when a NextGen step fails or has to retry, the agent now surfaces what happened inline. Failed steps appear with a recovering (orange) or terminal (red) state, and a "What happened?" accordion explains the failure in plain language. The Actions & tasks block can also be collapsed.
Document Analyst card unified - the separate Doc Analyst and KB Doc Analyst cards have been merged into a single Document Analyst card. In the Configure modal, you can now choose whether the agent analyzes user-attached files, the agent's knowledge base documents, or both. No functionality change - just less to set up. Applies to both legacy and NextGen agents.
Vision processing for image pages in the knowledge base - when Document Analyst targets a knowledge-base page that's an image, the agent now runs a fresh vision pass on the original asset (with the user's question and conversation context) instead of relying on stored OCR. Unlocks deep, in-question analysis of complex schematics, diagrams, and other image-heavy documents stored in the agent's knowledge base.
TODO.md no longer treated as a real file - the planner and replanner now recognize the [**TODO.md**] marker (used by the UI to render a clickable to-do chip) as a UI element, so the agent no longer wastes steps trying to look it up in the knowledge base.
🛠 Improvements
PDF Viewer demo from Citations settings - a new "Play PDF viewer demo" action in Chatbot Settings → Citations opens an in-context preview of the PDF citations experience.
Persona character limit shown live - the persona editor now shows the over-limit error as you type and disables Save until you're back under the limit, instead of waiting for a server validation error on save.
Cleaner Ask page title - the page title now reads Ask • [Agent Name] (centered dot separator) and no longer duplicates the agent name in the heading.
End-user conversation history is smoother - if a teammate starts chatting as a guest and then logs in, those earlier conversations now carry over to their account. The history panel slides in cleanly and stays open when you switch between conversations, and subdomain share-links now keep the history panel accessible when history is enabled.
Hybrid search auto-enables with Advanced Numeric Search - turning on Advanced Numeric Search at agent creation (or via the patch endpoint) now automatically enables hybrid search, since Advanced Numeric Search depends on it.
Claude 3.5 Sonnet has been retired. All agents previously using Claude 3.5 Sonnet have been automatically migrated to Claude 4.6 Sonnet. No action needed.
🔌 API & MCP
API reference rewrite across Tasks, Sources, Pages, and Conversations - cleaner descriptions, consistent NextGen terminology, per-value enum documentation for agent_capability, crawl_status, index_status, data_refresh_frequency, and format, plus accurate behaviour for poll-once-then-404 task results, file size limits on POST /sources, and the labels[] filter syntax on conversation send-message.
is_advanced_numeric_search removed from the public API - this setting is web-only by design. It is no longer accepted on POST /api/v1/projects or exposed in OpenAPI schemas.
🐞 Bug Fixes
Fixed the issue where the Smart Tasks Configure modal save button silently did nothing. Settings now save and the success toast appears.
Fixed the issue where editing an already-enabled tool's configuration (e.g. tweaking a custom action) returned a "max enabled tools" error if you were at your plan's limit. The limit now only applies when actually enabling a new tool.
Fixed the Confluence spaces selector showing a broken-image artifact when an external space icon failed to load. The fallback stack icon now renders cleanly.
Fixed Multi-Agent appearing in the Quick Start section for brand-new users. Multi-Agent only makes sense when you have existing agents, so it's now hidden until you do.
Fixed the post-subscription redirect sending active teams to the demo-agent screen after re-subscribing. Active teams now stay on the normal billing flow; only truly empty dashboards get the onboarding redirect.
Fixed the Conversation Insights agentic-actions filter showing raw internal keys (e.g. expensive_model_claude-4.6-sonnet-thinking) as filter labels. Expensive-model usage now displays as Added Cost: Claude 4.6 Sonnet Reasoning.
Fixed the NextGen file manager auto-opening on mobile as soon as the first file in a conversation was generated. It no longer pops open on small screens.
Fixed an edge case on Claude where the NextGen agent could re-process the user's message instead of progressing through the plan.
Smart Tasks billing now tracks actual agent work - the credit cost of each Smart Tasks run scales with how much the agent actually read and produced, instead of a flat per-tool-call charge. Lightweight runs cost less; heavy runs (large files, long generated code) cost what they should.
Smart Tasks breakdown in the cost popup - the NextGen badge popup now shows Smart Tasks as a collapsible row that opens into three sub-rows: Task setup, File parsing, and Code generation. Users can see exactly which stage drove the cost on any Smart Tasks run.
Cleaner dashboard rendering - switched the dashboard chart library to a lighter, canvas-based renderer. Charts no longer clip inside the chat bubble, legends sit cleanly below the plot, and resize behavior is more predictable. Dark-background regressions are gone; charts default to a clean white background.
Tighter agent/UI naming for dashboards - the agent now refers to dashboards by their title in chat, matching what shows up in the file manager.
First-time NextGen enable preserves NextGen defaults - turning NextGen on for the first time on an existing agent no longer reverts max_actions_per_query to 1 or carries over stale legacy settings. The optimal NextGen preset is applied cleanly.
Fewer duplicated steps and empty-prompt errors on Anthropic models - hotfix for NextGen agents using Claude where the agent could repeat execution messages or send an empty prompt mid-task.
Deep document analysis fan-out fix - resolved a race condition where parallel deep-dive analysis on multiple files could miss one of the files. Deep dive now consistently fans out to every requested document.
🐞 Bug Fixes
Fixed the issue where some columns in uploaded CSV files (e.g. url, post_date) were silently ignored on newer agents. All column data is now preserved end-to-end during ingestion.
Multi-Agents are now available - combine several specialized agents under one chat, and let end-users pick which one they talk to.
New Multi-Agent card on the Create page.
Users can build a Multi-Agent by attaching child agents to it.
Child agents can be added or removed at any time on the Multi-Agent's build page.
Rule of thumb: Multi-Agent controls UI, child controls AI.
When interacting with a Multi-Agent for the first time, the end-user is asked which child they want to chat with.
Users can switch the child they're talking to at any time, even mid-conversation.
Available on all deployments except SGE.
Available on Premium and Enterprise.
🤖 NextGen Updates
Generate Dashboard action - NextGen agents can now produce interactive dashboards (dynamic charts, filters, formulas) from whatever data is in the conversation: uploaded files, knowledge-base results, or numbers computed by the agent. The planner invokes this automatically when data is best presented visually, or on direct user request. Controlled by a new permission under Smart Tasks.
Side-panel artifact viewer - clicking a generated file in the file manager now opens it in a preview panel alongside chat. Supports markdown, plain text, images, and live dashboards. No download needed.
Full cost transparency on every NextGen message - the NextGen badge in the bottom-right of any response opens a popup with a full breakdown of the query steps and the exact credit cost. Same breakdown visible on Customer Intelligence pages.
Extended file type support for Smart Tasks uploads - when Smart Tasks Read Files is enabled, end-users can upload xlsx, xls, csv, tsv, json, md, and pptx alongside the usual document formats, so the agent can read and analyze them directly.
Conversation memory - NextGen conversations automatically retain earlier context: decisions, files, numbers, preferences - even 20+ messages later. Scoped per conversation; new conversations start fresh.
Smarter, more reliable agent behavior:
Smart Tasks carries memory of previous code-execution steps across the conversation and always returns a final summary.
Deeper document analysis works reliably again after a round of fixes.
The agent no longer hallucinates file names or entity references when re-planning mid-task.
Re-planning is more efficient, so complex NextGen messages use fewer credits.
NextGen polish:
The NextGen badge now has a proper hover state.
NextGen agents with no knowledge base can now hold a conversation, and past conversations are visible.
Deep document analysis is skipped automatically when the agent has no indexed pages.
Toggling NextGen on or off now saves cleanly, with no stale settings left behind.
The API error returned when NextGen is disabled now names the exact setting to enable.
When NextGen is off, Knowledge Base appears in the locked tools section alongside the other NextGen-only tools.
🛠 Improvements
Image citation limit per response - new setting in Personalize > Citations controls how many image or OpenGraph citations are rendered inline per response. Options: 1, 2, 3, or All. Extras beyond the limit show as regular numbered citations - nothing is lost, just less clutter on image-heavy answers.
Model selector upsell - users without model-selection access can now open the dropdown and see every model listed with a lock icon. Clicking any locked model opens a Contact Sales upsell explaining that Enterprise unlocks frontier models from OpenAI, Anthropic, and Google.
MCP / custom actions limit configurable per Enterprise plan - the per-agent cap on MCP and custom actions is now adjustable per Enterprise plan on request. Default of 3 stays on all other plans.
Instant Viewer marked as legacy - the Instant Viewer integration card now shows a Legacy badge with a tooltip pointing to the newer PDF Citations feature in Agent Settings.
OneNote files display as files in the SharePoint and OneDrive file pickers, not folders.
🔌 API & MCP
Page search - GET /api/v1/projects/{projectId}/pages now accepts an optional search query parameter that filters pages by URL or filename (case-insensitive substring match). OpenAPI spec updated.
Clearer NextGen API error - when use_nextgen_agentic_flow is disabled, POST /projects/{id}/tasks and POST /projects/{id}/tasks/{taskId}/messages now return a message naming the exact setting to enable.
🐞 Bug Fixes
Fixed the issue where the Ask Me Anything conversation would unexpectedly scroll to the bottom when image citations shuffled in an older message.
Fixed the issue where HTML entities (e.g. ', &) appeared as literal text in the conversation list sidebar and on Analyze and Ask widgets.
Fixed the issue where Instant Viewer opened for non-PDF inline citations (it should only open for confirmed PDFs; other citations now open as external links).
Fixed the issue where the PDF preview panel could carry stale state across conversations, showing the wrong document or highlight when switching chats.
Fixed the issue where the Instant Viewer legacy badge tooltip incorrectly showed a "Chat to learn more" button.
Fixed the issue where the CustomGPT Copilot send button stayed active on empty or whitespace-only input, and the input's resting and active states were inverted.
Fixed the Labels editor: the Delete All Labels button no longer throws a route error, and the add-label form is no longer artificially capped in width.
Fixed the Copied! tooltip in copy-to-clipboard controls so the confirmation appears in the right place.
End-user Conversation History (Premium) - personal conversation history is now available on all agent deployments (except SGE).
Team members: conversations sync across devices and browsers, persist indefinitely unless the user deletes them.
Guests: conversations save to browser storage - lost if the cache is cleared or the device changes.
Vimeo integration is now live as a native agent source.
Ingest Vimeo channels, user profiles, showcases, and individual videos.
Citations include timestamps that deep-link into the exact moment in the video.
Auto-sync on Premium plans.
Video must have transcription available for sync to work.
🤖 NextGen Updates
API support for NextGen agents - full API parity with UI.
Tasks: GET /projects/{id}/tasks (list), POST /projects/{id}/tasks (create), GET /projects/{id}/tasks/{taskId} (poll), PUT /projects/{id}/tasks/{taskId} (rename), DELETE /projects/{id}/tasks/{taskId} (delete), GET /projects/{id}/tasks/{taskId}/export (CSV/PDF export).
Messages: POST /projects/{id}/tasks/{taskId}/messages (send follow-up), GET /projects/{id}/tasks/{taskId}/messages (list), GET /projects/{id}/tasks/{taskId}/messages/{promptHistoryId} (fetch single).
Files: GET /projects/{id}/tasks/{taskId}/messages/{promptHistoryId}/files (list), GET …/files/{fileId}/download (download artifact), DELETE …/files/{fileId} (delete artifact).
NextGen toggle also available via PUT /projects/{id}/settings using use_nextgen_agentic_flow.
Polling model: GET /tasks/{taskId} returns processing + chunk events while the session runs, then completed + remaining events on first read after the session ends. Rate-limited via a separate tasks.poll bucket so polling doesn't consume the general API quota.
NextGen badge with queries breakdown - NextGen messages now show a unified bolt + "NextGen" badge. Hovering the badge reveals a popup with per-tool task counts and the actual query cost. Also visible on Customer Intelligence pages.
Chat with a NextGen agent that has no knowledge base - NextGen agents that rely only on tools (web search, custom actions, MCP) with no indexed documents can now chat. Previously these agents were blocked by a "add knowledge to your agent" gate.
Actions page polish on first NextGen activation - five sensible default actions (Knowledge Base, Smart Tasks, Doc Analyst, KB Doc Analyst, Web Search) are auto-enabled the first time NextGen is turned on. Incompatible actions render fully greyed out below the NextGen banner so what's available is obvious at a glance.
Knowledge Base Doc Analyst: multi-page search - Doc Analyst can now search and analyze multiple pages in one query instead of one at a time.
Agent accuracy and reliability improvements:
Cleaner response formatting - agents no longer occasionally leak stray code or markup into chat answers.
More reliable coding tasks when using Anthropic models (no more hangs).
Smarter tool selection - the agent is more aware of which tools are enabled and passes parameters through them more reliably.
More accurate answer classification feeding into Customer Intelligence metrics.
NextGen bug fixes:
Knowledge Base Doc Analyst and hybrid search no longer fail intermittently.
GPT-4.1 removed as a coding-task model (it was causing failures).
Max tasks per query is no longer reset to the plan default when NextGen is re-toggled - the default is set only on the very first activation.
🛠 Improvements
Customer Intelligence: Agentic Action filter - filter the Analytics Overview by prompts that used a specific agentic action.
Agent Initiative: don't prompt on mobile (Premium) - new Live Chat setting prevents agent-initiative starter-question prompting on mobile devices.
Azure provider flexibility for GPT models - several GPT models (GPT-4.1, GPT-4.1-mini, GPT-4o-mini, GPT-5.1, GPT-5.2, o4-mini variants) no longer hardcode OpenAI as the provider, so Azure routing works when selected at the project or plan level.
Chat input placeholder fade - the Ask Me Anything input now fades long placeholder text at the right edge instead of awkwardly clipping or causing horizontal scroll.
🔌 API & MCP
Multi-file upload - POST /projects/{id}/sources (plus POST /projects and PUT /projects/{id}) now accepts a files[] array of up to 50 files per request, matching the UI. The existing singular file input remains as a legacy alias for backward compatibility.
Source Labeling: AND/OR label grouping - nested AND/OR logic on source labels is now supported via API, letting callers express complex access rules.
🐞 Bug Fixes
Fixed the issue where Source Labeling with grouped labels returned "I'm sorry" responses on pages the user had legitimate access to.
Fixed the issue where users with zero source labels could incorrectly see all pages when Source Labeling was enabled, instead of the expected empty-set enforcement.
Fixed the issue where SharePoint and OneDrive folder changes failed with Undefined array key "cTag" errors during sync.
Fixed the issue where creating a new conversation did not appear in the sidebar until page refresh when the project's User Conversation History setting was Hidden or Team-Members-only.
Fixed the issue where YouTube citations lost their timestamp deep link when the video was referenced via a starter question.
Fixed the issue where large photo-heavy PDFs hung during upload.
Fixed the issue where Azure-hosted reasoning and GPT-5 deployments returned Unsupported parameter: 'max_tokens' errors.
