Changelog

New features, improvements, and fixes shipped to PixTwins.

NEWv2.0.84April 29, 2026

Install PixTwins on your phone

PixTwins now installs to your home screen on iPhone and Android, runs full-screen like a native app, and can push notifications to you when an agent finishes a long task or needs your approval. Every primary screen has been tuned for thumb-only use.

  • -Add to Home Screen on iPhone (Safari → Share → Add to Home Screen) or Android (Chrome → menu → Install app)
  • -Push notifications: get pinged on this device when an agent finishes work or needs you
  • -Chat sidebar tucks behind a hamburger on phones; composer respects the iOS safe area
  • -Agents, Memory, Outputs, Settings, and the marketing site all stack and wrap cleanly at 360–390 px
  • -Pinch-zoom is now allowed (accessibility) and tap targets are at least 44 px on small screens
NEWv2.0.83April 29, 2026

Browse 30+ agency-grade templates and install in one click

A new Templates page surfaces the full catalog of prebuilt twins and bundled teams. Pick a template, hit install, and you land in a fresh chat with your agent already onboarded — no instructions to write, no setup to do.

  • -New /templates page with search and category sections (Marketing & Growth, Content & Creative, Engineering & Data, Operations & Strategy, Quality Assurance, Video Production)
  • -Detail page per template shows what the twin does, how to talk to it, the tools it uses, and the integrations it pairs best with
  • -One-click install adds the agent to your workspace, seeds a starter chat with a friendly intro, and opens it for you
  • -Bundle install spins up a coordinated team (e.g. a full social content team) plus the brand and workflow memories they share — in a single click
  • -Empty /agents view now points new workspaces straight at the template catalog
BREAKINGv2.0.82April 29, 2026

PixTwins is now paid-only

PixTwins now requires an active paid plan. The free tier has been retired. All paid plans continue to include BYOK (bring your own provider key), so power users with their own Claude / Gemini / OpenAI key can still keep usage costs low while paying just the platform fee.

  • -Every workspace now requires an active Starter, Pro, or Max subscription
  • -BYOK (Anthropic / Gemini / OpenAI) remains available on every paid plan
  • -Subscriptions in past_due / canceled / unpaid status now return a clear "fix your billing" message rather than silently downgrading
  • -Existing paid customers: nothing changes for you
FIXv2.0.81April 29, 2026

More accurate billing windows + refund handling

Usage now tracks against your actual billing period instead of the calendar month, so mid-month signups get exactly one full allowance per cycle. Full refunds correctly end access on the refund date.

  • -Token + USD allowance windows now align with your subscription period (no more two partial windows for mid-month signups)
  • -Full Stripe refunds now end access immediately and accurately
  • -Out-of-order billing events from Stripe are ignored if they're older than what we already processed
FIXv2.0.80April 29, 2026

PixTwins CLI 0.2.1 — cleaner sign-out errors for scripts

Patch release: when the CLI is run while signed out, it now exits with code 1 and reports an authentication error, instead of looking like a server bug. Update with `npm install -g pixtwins@latest`.

  • -Signed-out commands now exit with code 1 (user error) and code AUTH_EXPIRED, not exit 2 (server error)
  • -Scripts can reliably distinguish "you need to log in" from "the server is down"
  • -Backstop heuristics map common error messages (permission denied, not found, network unreachable) to their proper exit codes
NEWv2.0.79April 29, 2026

PixTwins CLI 0.2.0 — full terminal interface

The pixtwins command-line tool now mirrors what you can do in the web app. Browse and edit agents, memories, integrations, automations, skills, and outputs from your terminal. Run `npm install -g pixtwins@latest` to upgrade.

  • -New commands: workspace, chats, agent, memory, integration, trigger, output, usage, billing, keys, automation, workflow, skill, capability, bundle, activity, report
  • -Inside the chat REPL: /help, /chats, /use, /workspace, /agents to switch context without leaving the conversation
  • -Pipe-friendly --json output on every command (auto-enabled when stdout is not a TTY) for scripting
  • -Bring-your-own-key management: pixtwins keys add/remove/test for Claude, Gemini, and OpenAI
  • -Open the Stripe billing portal directly from the terminal: pixtwins billing
FIXv2.0.78April 29, 2026

PixTwins CLI: login fix on current Chrome and Edge

Fixes a regression where signing in to the `pixtwins` command-line tool would silently fail on the latest Chrome and Edge. Run `npm install -g pixtwins@latest` to pick up the fix; existing sessions keep working.

  • -Browser-based `pixtwins login` now completes on Chrome 117+ and current Edge
  • -No action required if you only use the web app
  • -CLI updates: `npm install -g pixtwins@latest`
NEWv2.0.77April 27, 2026

All frontier models on every plan

Every PixTwins plan now leads with access to the most powerful models from Claude, Gemini, and OpenAI -- one subscription, every frontier model, no separate per-provider accounts. The pricing page, landing page, and in-app model picker have all been updated to make this front and centre.

  • -Pricing now leads with model access on every plan, not just token allowances
  • -New "frontier models" section on the landing page explains when to pick each provider
  • -Comparison table now shows multi-model choice as a PixTwins-only capability
  • -In-app model picker promoted to the chat header for easier mid-task switching
  • -New /model slash command and inline tip make per-turn provider overrides discoverable
NEWv2.0.76April 26, 2026

New plans, monthly token allowance, and priority support

Plans rebuilt around a monthly token allowance you can spend across Claude, Gemini, or OpenAI. Heavier models burn more tokens, lighter ones burn fewer — same pool, your call. Bring your own API key on every plan. Pro now includes priority support; the new Max plan includes expert setup and a custom-branded portal.

  • -Starter $99/mo includes 4.95M tokens of monthly AI usage across providers
  • -Pro $399/mo includes 20M tokens of usage and priority support for failures and quality issues
  • -Max $999/mo replaces Team — 50M tokens of usage, audit + human expert setup, custom-branded portal and apps for your business
  • -Bring your own API key (Claude, Gemini, or OpenAI) on every plan, including free
  • -Mix and match providers freely — heavier models cost more tokens, cheaper ones cost less
  • -Free tier now allows your own Anthropic key to unlock the most powerful model
NEWv2.0.75April 26, 2026

Choose your AI provider — Claude, Gemini, or OpenAI

Switch the engine that powers your workspace between Claude, Gemini, and OpenAI. The chat header always shows which model is running, and the provider selector now lives next to the chat box.

  • -Three-way provider toggle now sits beside the chat input on every conversation
  • -OpenAI mode runs on the GPT-5.5 family with adjustable reasoning depth
  • -Per-message overrides via slash commands (/openai, /gpt5.5-fast, /gpt5.5-deep) for one-off turns
  • -Active model badge in the chat header now labels every provider, not just Claude
FIXv2.0.74April 26, 2026

Faster image generation

Image creation now starts with the faster image generation path for everyday creative work, while keeping a high-quality fallback when needed.

  • -Image requests now use the faster generation path first
  • -A higher-quality fallback is still available automatically if the first path is temporarily unavailable
  • -Removed an outdated fallback that could cause some image requests to fail
NEWv2.0.73April 26, 2026

Choose your AI provider — Claude, Gemini, or OpenAI

Switch the engine that powers your workspace between Claude, Gemini, and OpenAI. The chat header always shows which model is running so you can pick the right one for the task.

  • -Three-way provider toggle in the top nav: Claude, Gemini, OpenAI
  • -OpenAI mode runs on the GPT-5.5 family with adjustable reasoning depth
  • -Per-message overrides via slash commands (/openai, /gpt5.5-fast, /gpt5.5-deep) for one-off turns
  • -Active model badge in the chat header now labels every provider, not just Claude
  • -Bring your own OpenAI key in Settings to bypass workspace token limits
FIXv2.0.71April 26, 2026

More reliable agent actions

Agents now choose compatible actions more consistently when you create a team, install a bundle, connect apps, or ask for code, data, media, and monitoring work.

  • -New agents are set up with action sets that match what PixTwins can actually run
  • -Connected Gmail, Slack, Calendar, Analytics, and Shopify actions are surfaced more clearly when available
  • -Agents avoid claiming unavailable app actions and give clearer guidance when a connection is missing
IMPROVEDv2.0.70April 25, 2026

Better connected app action loading

When you connect an app like Klaviyo, Gmail, or Slack, your agents now receive a better-balanced set of available actions for the current task. Bundle-installed agents can also pick up relevant new app connections later instead of being locked to the apps the bundle was originally built around.

  • -Agents see a broader, task-aware action set from connected apps
  • -Bundle-installed agents can adapt to relevant new connections you add later
  • -Very large tool responses are safely shortened with a clear note so your agent can ask you to refine the request
FIXv2.0.69April 22, 2026

Smoother chat on message send

Sending a message no longer makes the chat area briefly disappear and reappear. The message list now stays stable through the whole send — from the moment you hit enter to the agent's first word — instead of blanking out and redrawing a few times along the way.

  • -No more flicker or black flash when sending a message
  • -Message list stays put while the agent starts streaming
  • -Artifacts and file previews stay mounted instead of re-rendering every few seconds
NEWv2.0.68April 21, 2026

Edit and resend chat messages

Made a typo, forgot a detail, or want to rephrase a request? Hover over any message you sent and hit "edit". Update the text and your agent replies to the new version — everything that came after the edited message is cleared so you get a clean rerun instead of stacking on top of the old thread.

  • -Edit button appears next to every message you sent
  • -Saving replaces the message and removes all later replies so the agent answers fresh
  • -Press Enter to resend, Esc to cancel
  • -Disabled while a reply is still streaming so you never interrupt a response mid-flight
NEWv2.0.67April 20, 2026

PixTwins in your terminal

Drive your PixTwins workspace without leaving the shell. Install the CLI, sign in through your browser, and you can run one-shot prompts, drop into an interactive chat, or pipe prompts into scripts and cron jobs — all using the same agents, memories, and integrations as the web app.

  • -Install: npm install -g pixtwins
  • -Sign in via browser — `pixtwins login` — works with Google or email accounts
  • -One-shot mode: `pixtwins "summarize yesterday's GA4 traffic"` streams the reply and exits
  • -Interactive REPL: run `pixtwins` to chat back and forth, with /new to start a fresh chat
  • -Generated images, PDFs, and code artifacts auto-save to ./pixtwins-outputs/
  • -Scripting-friendly: `--json` emits structured JSONL you can pipe into jq or other tools
NEWv2.0.66April 20, 2026

Agents can now generate with fal.ai

Once your fal.ai key is in Settings, your agents get three new generation tools. Ask for an image and they can call Flux; ask for a video and they reach for Kling, LTX, Wan, or Veo 3; ask for a soundtrack and they call Stable Audio or MiniMax Music. Outputs land in your Outputs tab automatically.

  • -New tool "Flux Image (fal.ai)" — flux/schnell for drafts, flux/dev balanced, flux-pro/v1.1 for hero shots, flux-realism for photoreal
  • -New tool "Video (fal.ai)" — text-to-video via Kling v2 master/pro/standard, LTX, Wan, Mochi, or Veo 3, 5–10s clips, 16:9 / 9:16 / 1:1
  • -New tool "Music (fal.ai)" — Stable Audio, MiniMax Music, or CassetteAI, 3–600s backing tracks
  • -Generated images, videos, and music are saved to your Outputs tab automatically
  • -Add these tools to any agent from the agent builder — works alongside the existing Gemini, Higgsfield, and ElevenLabs tools
NEWv2.0.65April 20, 2026

Bring your own fal.ai key

Add your fal.ai API key in Settings to plug into hundreds of open models for image, video, audio, and 3D generation. Sits alongside your existing Anthropic, Gemini, Higgsfield, and ElevenLabs keys.

  • -Paste your fal.ai key under Settings → API Keys
  • -Key is encrypted at rest and validated when you save it
  • -Unlocks access to fast image/video/audio/3D generation models via your own fal.ai account
  • -Set once globally or per workspace, same as the other provider keys
IMPROVEDv2.0.64April 19, 2026

Sharper image generation

Our image tool now runs on a newer model with crisper text rendering and more accurate scene composition. Posters with headlines, product mockups, and any prompt that asks for text-in-image should look much cleaner now.

  • -Significantly better rendering of text inside images
  • -More faithful scene composition for complex prompts
  • -Automatic fallback to the prior model if the new one is temporarily unavailable
NEWv2.0.63April 19, 2026

Pick your engine — Claude or Gemini

Each workspace can now choose which provider runs the main agent loop. Use the toggle in the top bar to switch between Claude (premium quality, our default) and Gemini (efficient mode with very long context). Your choice persists per workspace.

  • -Top-bar toggle switches between Claude and Gemini modes per workspace
  • -Gemini mode is ideal for long-context research and high-volume work
  • -You can also force a single turn onto the other engine with /opus, /sonnet, /haiku, or /gemini, /gemini-flash
  • -All your tools, memories, integrations, and chat history work identically in both modes
FIXv2.0.62April 19, 2026

Shopify — fix store not linking when launched from Shopify admin

Opening the PixTwins app from inside your Shopify store admin now correctly connects the store to your workspace. Previously the OAuth round-trip would finish but silently fail to save the connection, leaving you on the chat page with no store linked.

  • -Logging in from the Shopify admin app launcher now links the store to your active workspace
  • -Connection from the Integrations page (enter-a-shop-domain flow) is unchanged
FIXv2.0.61April 19, 2026

Integrations — cleaner Shopify connect flow

Fixed a UI glitch on the Integrations page where the Shopify connect option could appear twice when other integrations were already connected.

  • -Shopify now shows a single connect card for your first store
  • -The inline "+ Add Shopify store" button only appears after at least one store is connected, for adding additional stores
IMPROVEDv2.0.60April 18, 2026

Billing Reliability Improvements

Hardened the subscription system to prevent duplicate charges and keep subscription status accurate during delivery hiccups.

  • -Duplicate subscription attempts are now blocked with a clear message directing you to the billing portal
  • -Subscription status updates are more reliable during temporary network issues
  • -Faster recovery from payment failures — your account unlocks as soon as payment succeeds
NEWv2.0.59April 18, 2026

Social Studio — Calendar view + direct publishing

Social Studio gets a month-grid calendar showing every scheduled post at a glance, and posts approved on the board can now actually go live. Connect the social integrations you use and hit publish — or let the scheduler handle queued posts automatically.

  • -Calendar tab on /studio/social with a full month view, today indicator, prev/next navigation, and a per-day drill-down
  • -Each day cell shows scheduled posts colour-coded by status — draft, scheduled, published, failed — with time and target platforms
  • -Direct publishing from the board — a "publish" button on approved/scheduled/failed cards sends to connected platforms immediately
  • -Automatic publishing for scheduled posts — the system checks every 15 minutes and posts anything that's due
  • -Per-channel outcome tracking — if one platform fails, the others still go out and the failure shows on the card
  • -Supported platforms for text posts: LinkedIn, X, Facebook, Threads, Bluesky. Instagram, YouTube, and TikTok (media-first platforms) coming next
NEWv2.0.58April 18, 2026

Social Studio — plan, draft, approve, and schedule social content from one board

Install the Social Studio bundle from the marketplace to unlock a purpose-built workspace for social media: a full team of specialist agents plus a Kanban post board covering every stage from idea to published.

  • -One-click install from the marketplace or onboarding — adds a social team (content manager, scheduler, approval coordinator, community responder, analyst) and seeds brand-voice memories
  • -New Social Studio tab with a Kanban board: Idea → Draft → In Review → Approved → Scheduled → Published
  • -Draft posts with per-channel captions, target platforms, and a scheduled time from one dialog
  • -Ask any agent in chat to draft or move posts — they appear on the board instantly
  • -Approval-gated flow — posts can't reach Scheduled until they're Approved
  • -Calendar, Inbox, and Analytics tabs marked for upcoming releases
NEWv2.0.57April 18, 2026

Your agents now have a persistent computer

Every workspace gets a real working environment where agents install libraries, log into sites, and run watchers that keep going between sessions.

  • -Files stay where you put them — your agents do not start from zero every conversation
  • -Agents can log in to third-party sites once and stay signed in
  • -Run background watchers and schedulers that survive across sessions
  • -A new Computer tab shows what is running, what has been created, and lets you reset if needed
NEWv2.0.56April 18, 2026

Subscription Plans & 7-Day Free Trial

Subscription plans are now active. Start with a 7-day free trial on any paid plan — no charges until the trial ends, cancel anytime. Manage your subscription from Settings.

  • -Starter ($29/mo), Pro ($119/mo), and Team ($209/mo) plans now available
  • -7-day free trial on every paid plan — full access, cancel anytime before day 7 with no charge
  • -Upgrade directly from the Settings page
  • -Manage, pause, or cancel your subscription from the Stripe billing portal
  • -Clear status indicators show trial days remaining and renewal dates
NEWv2.0.55April 18, 2026

Always-On Setup Wizard

Creating an always-on automation is now a guided 4-step flow. Pick the trigger type, configure it, describe the action, and launch — your agents handle the rest.

  • -Step-by-step wizard guides you through trigger type, schedule/event, action, and naming
  • -Four trigger types: Schedule, App Event, URL Monitor, Webhook
  • -Preset schedules (daily morning, Monday morning, every 6 hours) and common app events (Gmail, Slack, Calendar, GitHub, Shopify, Stripe)
  • -Review summary before launch so you know exactly what will happen
  • -Generates a structured setup message so agents wire up the automation end-to-end
IMPROVEDv2.0.54April 18, 2026

Landing Page FAQ

Added a common-questions section to the landing page so visitors can quickly get answers about what PixTwins does, pricing, data safety, and team workspaces without having to email support.

  • -Eight high-value FAQ items answering the most common questions
  • -Collapsible lo-fi accordion keeps the page uncluttered
  • -Quick FAQ link in the footer for direct navigation
IMPROVEDv2.0.53April 18, 2026

Smarter Agent Mentions

Mentioning a specific agent in chat is now faster and clearer. You can see which agent will receive your message before you send it, quickly reply to any agent with one click, and mention detection is more accurate.

  • -"@ reply" button on every agent message — one click prefills the mention in your input
  • -"Routing to @AgentName" indicator shows which agent will respond before you send
  • -Mention popup shows a green status dot for each active agent and a keyboard hint
  • -Mention detection no longer gets confused by email addresses or partial name matches
NEWv2.0.52April 18, 2026

Workspace Team Members

Invite collaborators to your workspace. Share access with teammates so they can view agents, chats, and memories without needing to set everything up themselves.

  • -Invite teammates by email directly from the Settings page
  • -Invited collaborators receive an email with a one-click accept link
  • -See all pending and active members in your workspace at a glance
  • -Remove members at any time from the Settings page
NEWv2.0.51April 17, 2026

Usage Charts, Scheduled Reports & Memory Search

The dashboard now shows 14-day token and task trend charts so you can see how your agents are being used over time. You can also schedule daily or weekly digest emails, and search through your memories instantly.

  • -Dashboard shows 14-day bar charts for token usage and task volume with week-over-week trend indicators
  • -Schedule daily or weekly digest reports — emailed automatically with task count, token usage, success rate, and top agents
  • -Memory page now has a live search bar — filter across all memories by label or content, with match highlighting
  • -Search shows a "no results" state when nothing matches your query
NEWv2.0.50April 17, 2026

Knowledge Graph Memories, Activity Log & Smarter Onboarding

Memories are now stored as a connected knowledge graph — relationships between facts are detected automatically, similar memories are grouped into communities, and context selection prioritizes the most relevant clusters. Plus a new Activity page and an improved onboarding wizard.

  • -Memory Graph view on the Memory page — visualize how your memories connect as an interactive node graph
  • -Memories are automatically linked with typed relationships (supports, contradicts, specializes, depends on)
  • -Related memories are clustered into named communities using graph analysis
  • -Agent context selection now prioritizes the most connected memory clusters for each query
  • -New Activity page — see every trigger event with outcome, retry count, agent response preview, and token usage
  • -Onboarding wizard now shows a step indicator so you always know where you are in setup
  • -Rebuild Team button on the Agents page lets you restart team setup at any time
NEWv2.0.49April 17, 2026

Custom Skills via Chat & Agent Editor Improvements

You can now ask any agent to create a reusable custom skill mid-conversation. The skill is saved to your workspace and automatically added to that agent. The agent builder also has a cleaner editing experience.

  • -Say "create a skill that…" in any chat — the agent builds and saves it instantly
  • -Custom skills appear in the Agents page with inline editing
  • -Edit agent triggers as tags — add with Enter, remove with Backspace
  • -Character count shown when writing agent instructions
  • -Composio toolkit configuration clarified — shows active status or specific pinned toolkits
NEWv2.0.47April 15, 2026

Visual Workflow Editor for Always-On Chats

Always-on chats now have a visual workflow editor that shows exactly what your automation does — trigger, actions, conditions, and delays — as a node graph you can view and adjust.

  • -See your always-on workflow as a visual node graph with connected steps
  • -Click any step to view and edit its configuration — field mappings, conditions, delays
  • -Add new steps between existing ones directly from the graph
  • -Drag nodes to rearrange the layout, positions auto-save
  • -View recent execution history with status, duration, and step progress
  • -Access from the "workflow" button on any always-on chat, or from Agents → Always Running tab
  • -Empty state when no workflow exists yet — just chat with your agent to create one
NEWv2.0.46April 13, 2026

Voice Mode with Full Agent Capabilities

Voice conversations now route through the full agent pipeline. Your agents can use tools, create automations, delegate to specialists, and save memories — all via voice.

  • -Voice mode now has access to all connected integrations (Gmail, Sheets, Slack, etc.)
  • -Create always-on automations and monitors by speaking
  • -Agents delegate to specialists during voice conversations just like text chat
  • -Long responses are automatically summarized for voice while full text is saved to chat
  • -Real-time status updates during processing — hear what your agent is doing
  • -Workspace memories are saved and retrieved during voice conversations
NEWv2.0.45April 13, 2026

Real-Time Voice Conversations

Talk to your agents in real time with the rebuilt voice conversation mode. Full-duplex audio — speak naturally and hear your agent respond instantly.

  • -Real-time bidirectional voice — no more waiting for text transcription
  • -Agent-aware conversations using your workspace context and memories
  • -Live transcript displayed during the conversation
  • -Full transcript saved to chat history when the session ends
  • -Voice mode now available immediately — no need to send a text message first
  • -Choose from 8 different voice options
NEWv2.0.44April 12, 2026

Full Shopify Read/Write + Order Fulfillment

Agents can now get individual product details, create new products, view and update orders with fulfillment tracking, and look up customer profiles.

  • -Get detailed product info including all variants, images, and tags
  • -Create new products directly from chat with customizable pricing and variants
  • -View individual order details with line items, fulfillments, and refunds
  • -Update orders: add notes, tags, and create fulfillments with tracking info
  • -Look up customer profiles with addresses, order count, and spend history
  • -Real-time webhook notifications for order, product, and customer events
NEWv2.0.42April 12, 2026

Notification Preferences & New Alert Types

Customize which notifications you receive with per-type preferences. New notification types for automation results and system announcements.

  • -Per-type notification preferences with in-app toggle directly in the notification dropdown
  • -New notification types: automation complete/error, always-on alerts, system announcements
  • -Widget notification count now includes app notifications alongside chat and trigger events
  • -Automatic cleanup of old notifications to keep things tidy
NEWv2.0.41April 12, 2026

Secure Share Links for Outputs

Outputs now use secure, token-based share links. Create multiple links per output, set expiration dates, track view counts, and revoke access anytime.

  • -Token-based share URLs -- output IDs are never exposed publicly
  • -Expiration options: no expiry, 1 day, 7 days, or 30 days
  • -View count tracking shows how many times each shared link has been accessed
  • -Revoke any share link instantly to cut off access
  • -New share modal with full link management
NEWv2.0.40April 12, 2026

Seamless Desktop Auto-Updates

The desktop app now checks for updates automatically and installs them in the background. No more manual downloads.

  • -One-click in-app updates -- no need to visit the website or re-download
  • -Update banner appears when a new version is available
  • -Signed updates for secure, tamper-proof delivery
NEWv2.0.39April 12, 2026

Marketplace Ratings & Featured Skills

Rate and review community skills in the marketplace. A new featured section highlights top-rated skills.

  • -Rate any imported skill from 1-5 stars with optional review text
  • -Featured section at the top showcasing the best community skills
  • -Sort skills by most imported, highest rated, or newest
IMPROVEDv2.0.38bApril 12, 2026

Persistent Chat Feedback

Your thumbs up/down feedback on agent messages now persists across page reloads. Feedback is stored per message and rate-limited.

  • -Feedback state survives page reloads
  • -Rate limiting prevents feedback spam (30 per minute)
  • -Feedback still drives agent performance scoring and memory confidence
NEWv2.0.38April 12, 2026

Voice Mode

Toggle voice mode in any chat to talk with your agent in real-time. Your agent listens, thinks, and speaks back instantly — a natural back-and-forth conversation.

  • -Real-time voice conversations with your AI agents — no more push-to-talk
  • -Choose from 8 natural voices for your agent's personality
  • -Interrupt your agent mid-response to redirect the conversation
  • -Automatic speech detection — just talk naturally, no buttons needed
  • -Voice minutes included in every plan (10 min free, 60 starter, 300 pro, unlimited team)
  • -Full transcript visible so you can follow along
NEWv2.0.37April 12, 2026

Notification Preferences & Expanded Alerts

Notifications now cover more events across the platform, with granular per-type preferences so you control exactly what shows up. The desktop widget also includes notification counts in its badge.

  • -New notification types for automation results, always-on alerts, and system announcements
  • -Per-type notification preferences -- toggle each notification type on or off from the bell dropdown
  • -Type badges on each notification show the category at a glance ([share], [auto], [alert], [system])
  • -Desktop widget badge now includes unread notification count alongside always-on and trigger counts
  • -Automatic cleanup of old notifications to keep your inbox tidy
NEWv2.0.36April 12, 2026

Full Shopify Read/Write + Order Fulfillment

Agents can now get individual product details, create new products, view and update orders with fulfillment tracking, and look up customer profiles. Expanded store management scopes for complete control.

  • -Get detailed product info including all variants, images, and tags
  • -Create new products directly from chat with customizable pricing and variants
  • -View individual order details with line items, fulfillments, and refunds
  • -Update orders: add notes, tags, and create fulfillments with tracking info
  • -Look up customer profiles with addresses, order count, and spend history
  • -Real-time webhook notifications for order, product, and customer events
NEWv2.0.34April 12, 2026

Notification Preferences & New Alert Types

Customize which notifications you receive with per-type preferences. New notification types for automation results and system announcements, plus automatic cleanup of old notifications.

  • -Per-type notification preferences with in-app toggle directly in the notification dropdown
  • -New notification types: automation complete/error, always-on alerts, system announcements
  • -Widget notification count now includes app notifications alongside chat and trigger events
  • -Automatic cleanup of old notifications to keep things tidy
NEWv2.0.33April 12, 2026

Secure Share Links for Outputs

Outputs now use secure, token-based share links instead of exposing raw IDs. Create multiple links per output, set expiration dates, track view counts, and revoke access anytime.

  • -Token-based share URLs -- output IDs are never exposed publicly
  • -Expiration options: no expiry, 1 day, 7 days, or 30 days
  • -View count tracking shows how many times each shared link has been accessed
  • -Revoke any share link instantly to cut off access
  • -New share modal with full link management -- create, copy, and revoke from one place
  • -Expired links return a clear "link has expired" message
NEWv2.0.32April 12, 2026

Seamless Desktop Auto-Updates

The desktop app now checks for updates automatically and installs them in the background. No more manual downloads -- just click "Update now" and the app restarts with the latest version.

  • -One-click in-app updates -- no need to visit the website or re-download
  • -Update banner appears when a new version is available with "Update now" and "later" options
  • -Proper version comparison so the app only prompts when a newer version actually exists
  • -Signed updates for secure, tamper-proof delivery
  • -New release tooling makes shipping updates faster for the team
NEWv2.0.31bApril 12, 2026

Marketplace Ratings & Featured Skills

Rate and review community skills in the marketplace. A new featured section highlights top-rated skills, and sort controls let you browse by popularity, rating, or recency.

  • -Rate any imported skill from 1-5 stars with optional review text
  • -Featured section at the top showcasing the best community skills
  • -Sort skills by most imported, highest rated, or newest
  • -Ratings and install counts update in real time
IMPROVEDv2.0.31aApril 12, 2026

Persistent Chat Feedback

Your thumbs up/down feedback on agent messages now persists across page reloads. Feedback is stored per message and rate-limited to prevent abuse.

  • -Feedback state survives page reloads -- no more losing your thumbs up/down
  • -Rate limiting prevents feedback spam (30 per minute)
  • -Feedback still drives agent performance scoring and memory confidence
NEWv2.0.31April 10, 2026

Pro UI Mode & Rich Chat

A major visual upgrade to the entire app. Chat messages now render rich formatting -- headers, bold, code blocks with syntax highlighting, tables, and more. Loading states are smoother, components are more polished, and keyboard accessibility is improved across the board.

  • -Rich markdown rendering in chat -- headers, bold, italic, lists, tables, and blockquotes display beautifully
  • -Syntax-highlighted code blocks with one-click copy buttons
  • -Smooth skeleton loading animations replace plain "Loading..." text across all pages
  • -New Pro/Classic UI toggle in Settings -- switch back to the original look anytime
  • -Improved keyboard navigation with visible focus indicators
  • -Toast notifications now work for copy, save, and error feedback
  • -Enhanced button, input, card, and badge components with proper loading, disabled, and error states
NEWv2.0.30April 10, 2026

Background Mode & Auto-Updates

The desktop app now stays running in your system tray when you close the window. Your widget and always-on agents keep working in the background. Plus, future updates install automatically.

  • -System tray icon with Show/Quit menu -- closing the window no longer kills the app
  • -Widget and always-on agents continue running in background
  • -Auto-updater checks for new versions on launch and installs them seamlessly
  • -In-app update notification for users on older desktop versions
NEWv2.0.29April 10, 2026

Floating Quick-Access Widget

Desktop app users get a sticky floating widget that stays on top of all windows. Chat or speak to your agents instantly — with full screen awareness, voice responses, and one-tap quick actions.

  • -Draggable pill widget — place it anywhere on your screen, always accessible
  • -Screen-aware: your agents can see what's on your screen and answer in context
  • -Voice in + voice out: speak your question, hear the answer
  • -Quick actions: one-tap Calendar, Email, and Screen analysis
  • -Notification badges for unread messages from always-on agents
  • -Global hotkey (Cmd+Shift+P) to toggle the widget
NEWv2.0.28April 10, 2026

Native Slack, Gmail, Calendar & Analytics

Slack, Gmail, Google Calendar, and Google Analytics now connect directly to your workspace — faster, more reliable, and no middleman. Agents get 24 new native tools.

  • -10 Slack tools: send messages, read channels & threads, search, react, upload files, list users
  • -6 Gmail tools: search inbox, read messages, send emails, reply in threads, create drafts
  • -5 Google Calendar tools: list events, create meetings, find free time, update & delete events
  • -3 Google Analytics tools: list properties, run reports, view real-time data
  • -Automatic token refresh for Google services — no manual reconnection needed
  • -Multi-workspace Slack support: connect multiple Slack teams to one workspace
NEWv2.0.27April 10, 2026

Multi-Store Shopify Support

Connect multiple Shopify stores to a single workspace. Manage all your stores from one place, set a default store for agents, and switch between stores when needed.

  • -Connect multiple Shopify stores to the same workspace -- no more separate workspaces per store
  • -Set a default store that agents use automatically when no store is specified
  • -Agents can target a specific store by name -- "check orders on my UK store"
  • -New "List stores" tool lets agents discover all connected stores and their status
  • -Add, disconnect, and manage individual stores from the Integrations page
NEWv2.0.26April 7, 2026

Auto-Plugin Discovery & Capability Registry

PixTwins now automatically detects what plugins, integrations, and tools your workflows need -- and helps you install them in one click. A unified capability registry brings together MCP servers, integrations, skills, and bundles into a single browsable catalog.

  • -Smart gap detection -- when you describe work or create agents, the system identifies what capabilities are missing and suggests what to install
  • -One-click bundles -- pre-configured capability packages (like "Video Production Pipeline") install everything you need at once
  • -Capability registry -- browse and search all available plugins, integrations, and skills from one unified catalog
  • -Agent awareness -- agents now know what capabilities are installed and will suggest missing ones instead of claiming access to tools they cannot use
  • -New suggest_capabilities tool -- agents can proactively detect and recommend missing capabilities during conversations
  • -Capabilities page -- manage installed plugins, review suggestions, and browse the full catalog from your workspace
NEWv2.0.25April 7, 2026

Persistent Context & Workspace Awareness

Agents now remember the full conversation history across long chats, and have awareness of other active work in your workspace. Stream reconnection is also more reliable.

  • -Rolling context memory -- agents maintain a persistent summary of your entire conversation, so nothing gets lost in long chats
  • -Workspace awareness -- each agent now knows what other chats are working on, enabling better cross-task coordination
  • -Smarter context management -- conversations are summarized incrementally instead of from scratch, saving cost and preserving nuance
  • -Better stream recovery -- if your connection drops, the app now resumes receiving progress in real-time instead of waiting for completion
  • -Structured task handoffs -- when agents need multiple rounds, the handoff context now includes decisions, tool results, and key findings
IMPROVEDv2.0.24April 6, 2026

Usage Tracking, Skills Polish & Reliability

See your token usage at a glance with the new usage badge, create and share skills from a community library, and benefit from stronger billing enforcement and expanded test coverage.

  • -Usage quota badge -- see your token consumption and budget in the nav bar, with color warnings at 80% and 95%
  • -Smarter billing enforcement -- budget exhaustion now returns a clear message instead of a generic rate limit error
  • -Community Skills library -- browse and install skills published by other users directly from the Skills tab
  • -Skill creation cards -- when agents create skills via chat, a visual card now appears inline in the conversation
  • -Skill input validation -- skill names and prompts are now validated for minimum length to ensure quality
  • -Expanded test coverage -- new automated tests for the chat API, orchestrator learning engine, and skill creation tool
FIXv2.0.23April 6, 2026

Artifact Persistence & Longer Agent Runs

Generated images, HTML previews, and other artifacts now reliably persist across page refreshes. Agents can also run longer on complex multi-step tasks without stopping prematurely.

  • -Artifacts survive page refresh -- previews and generated files are now linked directly to messages instead of matched by timing
  • -Wider recovery window -- existing artifacts are matched more reliably even for long-running generation tasks
  • -Longer agent runs -- complex multi-tool tasks (like generating multiple images) can now auto-continue up to 8 times instead of 5
FIXv2.0.22April 6, 2026

Security Hardening & Reliability Improvements

Comprehensive security audit and fixes across the platform. Stronger input validation, improved webhook protection, better error handling for long-running tasks, and startup health checks ensure your data and workflows are safer than ever.

  • -Stronger input validation -- message size limits and file name sanitization prevent abuse and protect system stability
  • -Improved webhook security -- replay attack prevention with timestamp verification ensures only legitimate events are processed
  • -Better billing protection -- payment failure handling no longer incorrectly reactivates canceled subscriptions
  • -Smarter file uploads -- workspace-to-chat ownership verification prevents cross-workspace data access
  • -Stream reliability -- 15-minute timeout on long-running streams prevents indefinite hangs
  • -Startup health checks -- critical configuration is validated at server startup, not at first request
  • -Safer background tasks -- atomic task creation prevents race conditions when sending rapid messages
NEWv2.0.21April 6, 2026

Shopify Agent Tools — Your AI Store Manager

Agents can now directly manage your Shopify store — browse products, check orders, create landing pages, edit theme templates, and analyze sales data. Connect your store and let your digital twin handle the day-to-day.

  • -Product management -- agents can browse, search, and update products, descriptions, tags, and pricing directly in your Shopify store
  • -Order & customer insights -- ask your agent about recent orders, customer data, and sales trends without leaving the chat
  • -Page creation -- agents can create and edit Shopify pages (landing pages, FAQs, about pages) with full HTML support
  • -Theme editing -- agents can read and modify Liquid templates, CSS, and JavaScript in your live Shopify theme
  • -Sales analytics -- get daily breakdowns of orders, revenue, and new customers for any date range
NEWv2.0.20April 5, 2026

Smarter Agents, Voice Mode & Multimodal Tools

Major intelligence upgrade — agents can now analyze images and screenshots, generate audio briefings, and respond to your voice. Plus significantly faster routing, upgraded web search, and reduced costs across the board.

  • -Upgraded web search -- faster, more accurate results with better source citations and broader coverage
  • -Voice Mode -- talk to your digital twin with push-to-talk voice input and listen to responses spoken aloud
  • -Image Analysis -- agents can analyze screenshots, documents, charts, and photos with AI vision
  • -Audio Overviews -- generate podcast-style audio briefings from any text content, reports, or research
  • -Faster routing -- intent classification and tool selection are now 3x faster with a more efficient routing engine
  • -Upgraded media generation -- image and video generation now use the latest generation models for higher quality output
NEWv2.0.19April 5, 2026

Auto-Recovery, Agent Lifecycle & Workspace Policies

Agents now automatically recover from transient failures instead of crashing. A new lifecycle system tracks agent state from start to finish, and workspace policies let you control agent behavior with declarative rules.

  • -Auto-recovery -- agents now automatically retry when they hit temporary issues like service overloads or rate limits, with smart backoff and escalation
  • -Agent lifecycle tracking -- every agent task now has a clear state (starting, running, paused, completed, failed) for better observability
  • -Workspace policies -- set rules to control agent behavior: model selection, budget caps, tool restrictions, and event routing, all evaluated instantly
  • -Recovery notifications -- see real-time updates when an agent is recovering from an issue, so you know what's happening behind the scenes
  • -Improved error messages -- clearer explanations when things go wrong, with context about what the system tried to do about it
NEWv2.0.18April 4, 2026

Background Knowledge Sync, Usage Tracking & Always-On Reliability

Agents now automatically learn from your connected apps. Gmail, Slack, and GitHub data is indexed in the background so your agents know about your work without being told. Plus, usage tracking, template upgrades, and always-on reliability improvements.

  • -Background knowledge sync -- agents automatically learn from your connected apps like Gmail, Slack, and GitHub, staying up to date without being told
  • -Usage tracking -- all activity now counts toward your plan budget, with clear limits for agents and memories per plan
  • -Template quality upgrade -- all templates now use specialist-grade instructions with real frameworks and methodologies
  • -Always-on reliability -- improved event processing with automatic retries and better error handling
  • -Knowledge memories -- a new memory category that stores context from your connected apps, searchable alongside your existing memories
NEWv2.0.17April 4, 2026

Deep Research, Smarter Reasoning & Memory Deduplication

Agents can now conduct multi-source research with parallel search and synthesis. They also reason more carefully through complex multi-step tasks, and workspace memories are automatically deduplicated in real-time.

  • -Deep research -- agents can now conduct thorough multi-source research and deliver structured reports
  • -Structured reasoning -- agents now think through complex decisions before acting, improving accuracy on multi-step tasks
  • -Real-time memory dedup -- workspace memories are checked for duplicates before saving, preventing bloat and keeping context clean
  • -Memory merging -- when new information overlaps with an existing memory, it gets merged instead of creating duplicates
  • -Research depth controls -- quick, standard, and comprehensive modes with plan-appropriate access
NEWv2.0.16April 3, 2026

Better Onboarding & Agent Performance Tracking

New users now get a smoother start with generalist agents, template previews, and smarter guidance. Plus, agent performance is now tracked and visible so you can see how your team is doing.

  • -Quick start option -- skip onboarding and get 5 ready-to-use generalist agents (Assistant, Writer, Researcher, Analyst, Developer)
  • -Template previews -- see exactly which agents will be created before committing to a template
  • -4 new templates -- SaaS Operations, E-commerce, Freelancer/Consultant, and Engineering Team
  • -Agent performance visible -- see success rate, task count, and average usage on each agent card
  • -Error tracking -- recent agent errors now appear on the dashboard so you can spot issues quickly
  • -Dynamic chat suggestions -- starter prompts now match your actual agent team instead of generic examples
IMPROVEDv2.0.15April 2, 2026

Smarter Context Memory & Faster Tool Loading

Agents now preserve structured context during long tasks, so they remember goals, progress, and key decisions when continuing work. Integration tools also load faster with parallel fetching.

  • -Structured context memory -- long conversations are summarized into organized sections so agents remember goals, progress, and key decisions
  • -Richer auto-continuation -- when a task needs multiple rounds, the agent now carries forward more detailed progress notes
  • -Faster integration loading -- connected app tools now load faster when you have multiple integrations
NEWv2.0.14April 1, 2026

Agent Instruction Version History & Rollback

Every time an agent learns and updates its own instructions, PixTwins now saves a versioned snapshot. You can view the full version history from the agent editor and roll back to any previous version with one click.

  • -Automatic versioning -- every self-learning update and instruction consolidation creates a new version with a summary of what changed
  • -Version history panel -- expand "Version History" in the agent editor to see all past versions, who changed them (you, self-learning, or consolidation), and when
  • -One-click rollback -- restore any previous version instantly if an agent learned something you disagree with
NEWv2.0.13April 1, 2026

Negative Learning (Anti-Pattern Memories)

When you correct an agent, the system now automatically saves what NOT to do as an anti-pattern memory. Agents learn from mistakes and actively avoid repeating them.

  • -Corrections like "no", "wrong", "don't do that" now trigger anti-pattern memory creation automatically
  • -Anti-pattern memories are injected into agent prompts with an "AVOID" prefix so agents clearly know what to steer away from
  • -New "Anti-Pattern" category on the Memory page with red-bordered cards for easy identification
  • -Anti-patterns make agents less likely to repeat failed approaches
  • -You can also manually create anti-pattern memories from the Memory page
NEWv2.0.12April 1, 2026

Agent Personality & Voice Settings

You can now configure each agent's communication style with tone, verbosity, and formality controls. Customize how your agents talk -- from concise and technical to friendly and detailed.

  • -Three new voice controls per agent: Tone (professional, casual, friendly, technical, concise), Verbosity (brief, standard, detailed), and Formality (formal, neutral, informal)
  • -Voice settings are injected into the agent's system prompt as a "Communication Style" section, shaping how it responds
  • -Agents can also set voice via chat -- just ask "make this agent more casual" and the tools handle it
  • -Default is no voice override, so existing agents behave exactly as before
IMPROVEDv2.0.11April 1, 2026

Progressive Tool Output Display

You can now see what each tool returned during agent execution instead of just "Using tool..." with no feedback until the final response arrives.

  • -Tool results now show a condensed summary as soon as each tool finishes
  • -Summaries display inline in the chat status area and auto-clear after 2 seconds
  • -Error results are clearly marked so you know immediately if a tool call failed
IMPROVEDv2.0.10April 1, 2026

Smarter Self-Learning: Instruction Pruning & Cross-Agent Knowledge

Agents now automatically consolidate their instructions when they grow too long from self-learning, keeping them focused and contradiction-free. Learnings are also shared across your team as workspace memories.

  • -Instruction pruning: when an agent's instructions grow beyond a threshold, they are automatically consolidated — merging duplicates, resolving contradictions, and removing generic advice
  • -Duplicate prevention: the reflection system now checks existing instructions before adding new ones, avoiding redundant self-learning
  • -Cross-agent knowledge sharing: when one agent learns something new, it is saved as a workspace memory so other agents on your team benefit too
NEWv2.0.9April 1, 2026

Smarter Agents: Adaptive Thinking, Feedback, Model Control & Learning

A major upgrade to agent intelligence and user control. Agents now think deeper on complex tasks, learn from your feedback, and can be individually tuned with model preferences and cost limits.

  • -Adaptive thinking: agents automatically reason more deeply on complex tasks and respond faster on simple ones
  • -Per-agent model control: assign each agent a preferred model tier (e.g. use the fastest model for simple lookups, the most powerful for code generation)
  • -Cost guardrails: automatic per-request budget limits based on your plan tier prevent unexpected costs
  • -Feedback buttons: thumbs up/down on any response — your feedback improves agent performance and memory confidence over time
  • -Improved learning: agents now save successful tool patterns to a knowledge base even without explicit corrections, making them better at recurring tasks
  • -Chat agent creation parity: create agents via chat with team grouping and integration access — same fields as the UI form
  • -Progress updates: see what subagents are doing during long-running delegated tasks
IMPROVEDv2.0.8April 1, 2026

UX Polish: Multiline Chat, Agent Teams, Workspace Rename & More

A quality-of-life update addressing the most-requested user feedback: multiline chat input, agent team grouping, direct agent chat, workspace renaming, improved empty states, better output visibility, and enhanced API key management.

  • -Multiline chat input: use Shift+Enter for new lines, auto-resizing textarea that grows as you type
  • -Agent teams: organize agents into named groups (e.g. CRO, Finance, Content) with visual grouping on the agents page
  • -Click-to-chat: start a conversation with any agent directly from their card on the agents page
  • -Workspace rename: edit workflow names from the workspace dropdown
  • -Improved empty agents page: clearer onboarding with step-by-step guidance and dual CTAs
  • -Better output visibility: widened artifact detection window and fallback messages when tools complete silently
  • -API key management: test keys to verify they work, key age tracking with rotation reminders, and global keys that apply across all workspaces
NEWv2.0.7March 29, 2026

Algorithmic Trader Workflow Template

New pre-built workflow for traders: a 5-agent team that covers the full trading lifecycle from market research to algo deployment. Includes real-time data tools, strategy backtesting, risk management, and automated trade execution.

  • -Market Analyst agent: real-time stock, crypto, and forex data with macro analysis and trade idea generation
  • -Strategy Backtester agent: Python-powered backtesting with Monte Carlo simulations, walk-forward analysis, and equity curve visualization
  • -Risk Manager agent: portfolio Greeks, exposure monitoring, stress testing, and position sizing validation
  • -Algo Engineer agent: production-grade trading bot development with broker API integration, paper trading mode, and kill switches
  • -Trade Journalist agent: automated trade logging, performance analytics, and PDF report generation
  • -Pre-loaded trading rules, strategy pipeline, and algo deployment checklists as workspace memories
  • -WorldMonitor.app integration for real-time market data feeds and custom alerts
NEWv2.0.6March 29, 2026

Dashboard, Marketplace, Gallery Upgrade & Financial Tools

Major platform expansion: agent performance dashboard with leaderboard, skill marketplace for community sharing, upgraded outputs gallery with search and bulk actions, plus real-time financial data tools for traders and analysts.

  • -Agent performance dashboard: team analytics, success rates, XP leaderboard with expandable agent detail cards
  • -Skill marketplace: publish your custom skills to the community, browse and one-click import skills created by other users
  • -Outputs gallery upgrade: search, sort (newest/oldest/name/size), grid and list views, bulk selection with batch delete
  • -Financial data tools: real-time stock/crypto/forex prices, economic indicators (VIX, yields, indices), and a full financial calculator with options pricing, CAGR, Sharpe ratio, and position sizing
  • -Navigation update: Dashboard and Marketplace now accessible from the top nav
NEWv2.0.5March 29, 2026

Bring Your Own Key: Use Your Own API Key

Your Anthropic API key now powers your entire workflow. Add your key in Settings and every agent call, routing decision, and tool execution uses your key instead of PixTwins credits. Work without token limits.

  • -End-to-end BYOK: every AI call in your workspace uses your own API key when configured — routing, agent work, summarization, tools, and self-learning
  • -No token budget limits: BYOK users bypass monthly token caps entirely since you pay the provider directly
  • -Free-tier unlock: BYOK users on the free plan get access to the most powerful model, same as paid plans
  • -Seamless fallback: remove your key anytime and your workspace automatically falls back to PixTwins credits
  • -Encrypted storage: your API key is encrypted at rest and never exposed to the frontend
NEWv2.0.4March 28, 2026

Custom Skills & Active Skill Prompts

Skills are now fully functional. When you assign a skill to an agent, its specialized instructions are injected into the agent's behavior. You can also create your own custom skills from the Skills tab, via chat, or inline in the agent editor.

  • -Active skill prompts: assigned skills now directly shape how your agent thinks and works, not just which tools it has access to
  • -Custom skill creation: build your own skills from the Skills Library with a name, category, and detailed prompt instructions
  • -Chat-based skill creation: describe a skill in chat (e.g., "create a skill that audits Shopify stores") and the agent generates it for you
  • -Inline skill creation: create and instantly assign a new skill from the agent editor without leaving the page
  • -Custom skills appear alongside built-in skills everywhere: the Skills Library, the agent editor, and the agent's active behavior
NEWv2.0.3March 28, 2026

Intelligence Upgrade: Smarter Routing, Learning & Memory

Major upgrade to the orchestration engine. Agents now route tasks to the right model based on complexity, track their own performance, automatically summarize long conversations, and consolidate related memories into higher-level insights.

  • -Smarter routing -- the system automatically matches each request to the right level of processing power, saving your budget for complex tasks
  • -Unlimited conversations -- long chats are automatically managed so they can run indefinitely without losing important context
  • -Agent health tracking -- each agent builds a performance profile so the system can prefer better-performing agents
  • -Memory consolidation -- related memories are automatically synthesized into concise insights, keeping your knowledge base focused and relevant
NEWv2.0.2March 26, 2026

Cross-Chat Search & PixTwin Identity

Agents can now search across all your past conversations in a workspace, finding relevant context from previous chats and memories. The main coordinator agent has been rebranded to PixTwin with an updated execution style.

  • -Cross-chat search: agents automatically find relevant context from your past conversations and workspace memories when answering questions
  • -PixTwin identity: the coordinator agent now introduces itself as PixTwin and plans before executing, giving you visibility into what it will do
  • -Streamlined memory: simplified memory categories for faster, more accurate recall
NEWv2.0.1March 25, 2026

DTC Email & Retention Agents

The Email Marketing template now includes three new specialized agents for DTC brands — a retention strategist that audits flows and plans A/B tests, a monthly performance reporter that pulls Klaviyo + Recharge data into Google Sheets, and a QA auditor that catches grammar errors and broken links weekly.

  • -Retention Strategist: on-demand flow audits, ranked optimization priorities (subscription conversion → churn reduction → revenue → list growth), structured A/B test briefs with copy and design direction
  • -Performance Reporter: monthly always-on agent that pulls subscription metrics from Recharge and email performance from Klaviyo, writes to Google Sheets with data interpretation and action items
  • -QA Auditor: weekly always-on agent that checks every flow and campaign for spelling, grammar, broken links, and link-copy mismatches — reports to Slack
IMPROVEDv2.0.0March 25, 2026

Agent Engine Upgrade

Major upgrade to the agent execution engine. Agents now run on a more capable foundation with smarter delegation, adaptive thinking effort, budget safety, and enhanced self-learning from your feedback.

  • -Smarter delegation: specialist agents now use your plan's most powerful model instead of a weaker one, and get more room to complete complex tasks
  • -Adaptive effort: the system automatically adjusts thinking depth based on task complexity -- quick answers for simple questions, deep reasoning for complex work
  • -Budget safety: per-query cost caps prevent any single task from consuming your entire monthly budget
  • -Enhanced learning: agents now learn aggressively from corrections and reinforcement -- say "perfect" or "no, try again" and they remember for next time
  • -Live delegation progress: see what specialist agents are doing in real-time during handoffs
  • -Faster execution: removed redundant planning step that was causing agents to announce intentions instead of acting
NEWv1.9.0March 25, 2026

Workflow Engine

Build structured "When X happens, Do Y" automations that execute without AI cost. Describe what you want in chat, and the agent guides you through trigger selection, field mapping, and confirmation with interactive UI elements -- all inline in the conversation.

  • -Structured workflows: trigger-based automations with action, condition, and delay steps
  • -Adaptive chat UI: inline dropdowns, field mapping grids, and confirmation cards inside agent messages
  • -Zero-cost execution -- workflows run without consuming your plan budget
  • -Smart error recovery -- when a workflow step fails, an agent analyzes the error and retries with corrected parameters
  • -New agent capabilities for building and managing workflows through chat
NEWv1.8.0March 22, 2026

Smarter Agent Engine

Upgraded our agent engine for more powerful autonomous workflows. Agents now have true agentic loops with tool use, error recovery, and multi-step planning built in.

  • -Agents now support multi-step planning with automatic error recovery
  • -Tool execution includes built-in retry logic for reliability
  • -Backward compatible -- existing agent configurations work without changes
NEWv1.7.2March 20, 2026

Trigger Pre-Filters

Always-on agents can now define pre-filters on their triggers so they only process events that match specific criteria. This reduces noise and saves your plan budget by skipping irrelevant events before they reach the agent.

  • -Filter evaluation happens before the agent is invoked, saving your plan budget
  • -Supports flexible matching rules including text patterns and nested data
  • -Configurable from within chat -- just tell your agent what to filter
FIXv1.7.1March 19, 2026

OAuth Redirect Fix

Fixed a bug where OAuth callbacks would redirect to the wrong URL after connecting an integration. OAuth redirects now work correctly in all deployment environments.

  • -Fixed OAuth redirect issues when connecting integrations
  • -Integration connection flow now works reliably across all environments
NEWv1.7.0March 18, 2026

Enhanced Integration Security

All integration credentials are now protected with additional security layers. Your connected app credentials are safer than ever with no changes needed on your end.

  • -All integration credentials are now protected with additional security layers
  • -No changes needed to existing workflows
  • -All previously stored credentials automatically upgraded
IMPROVEDv1.6.2March 16, 2026

Script Outputs in Chat

Code execution results now stream directly into the chat interface. Results are displayed inline so you can see exactly what your scripts produce without leaving the conversation.

  • -Code execution results stream into chat messages in real time
  • -Syntax-highlighted output blocks for better readability
  • -Error output distinguished with a different border color
  • -Large outputs are automatically truncated with a "show more" expansion
NEWv1.6.1March 14, 2026

Update Automation Tool

Agents can now modify their own automation configurations through a new tool call. This means you can ask an agent to change its schedule, update trigger filters, or adjust its behavior without visiting the settings page.

  • -Agents can now modify their own automation settings through chat
  • -Supports changing cron schedules, trigger filters, and instructions
  • -Changes are validated before applying to prevent invalid configurations
  • -Audit log tracks all automation modifications with timestamps
NEWv1.6.0March 12, 2026

Multi-Chat Architecture

Workspaces now support multiple independent chat threads. Create task chats for one-off work or always-on chats for continuous monitoring. Each chat maintains its own conversation history while sharing workspace-level agents and memories.

  • -Sidebar with chat list, create new, and switch between conversations
  • -Two chat types: task (one-off) and always_on (24/7 monitoring)
  • -Each chat has independent message history and continuation state
  • -Shared workspace agents, memories, and integrations across all chats
IMPROVEDv1.5.1March 10, 2026

Context Summarization

When a conversation grows very long, the system automatically summarizes earlier context to keep things running smoothly. This keeps conversations functional indefinitely without losing important details from earlier in the thread.

  • -Automatic summarization triggered when conversations grow large
  • -Structured summaries preserve key decisions and context
  • -Conversations resume quickly without re-processing earlier context
  • -Seamless to the user -- no interruption in the conversation flow
NEWv1.5.0March 8, 2026

Always-On Agents

Introducing 24/7 monitoring agents that stay active and process real-time events from your connected apps. Set up a Slack monitor, Gmail watcher, or analytics alerter that works around the clock without manual prompting.

  • -Always-on chat type with green pulsing indicator
  • -Real-time event triggers deliver events from your connected apps
  • -Agents process incoming events autonomously with full tool access
  • -Configurable via natural language -- just describe what to watch for
IMPROVEDv1.4.2March 6, 2026

Auto-Continuation

Long-running agent tasks now automatically continue where they left off when they reach a natural breakpoint. A continuation summary is generated and the next request fires seamlessly so complex tasks run to completion.

  • -Automatic continuation for long-running agent tasks
  • -Structured continuation summaries preserve progress between segments
  • -Seamless auto-resume with no manual intervention required
  • -Supports extended multi-part execution for very long tasks
NEWv1.4.1March 4, 2026

Code Execution

Agents can now write and execute Python and Node.js code in a secure environment. Each execution is fully isolated with automatic cleanup for safety.

  • -Supports both Python and Node.js
  • -Fully isolated and secure execution environment
  • -Automatic cleanup after every run
  • -Output captured and streamed back to the chat in real time
NEWv1.4.0March 2, 2026

Video Production Pipeline

A multi-agent workflow for end-to-end video creation. From research and brief writing through storyboarding, asset generation, and final editing -- your agent team collaborates to produce complete video content.

  • -Research agent gathers context and references for the video topic
  • -Brief and storyboard agents structure the narrative and visual plan
  • -Generation agent creates assets using AI image and video models
  • -Editing agent assembles the final video with Remotion
IMPROVEDv1.3.0February 28, 2026

Per-Agent Tool Access

Each agent now has its own set of allowed toolkits. This means your CRO agent only sees analytics tools while your email agent only sees Gmail -- reducing confusion and improving accuracy.

  • -Per-agent toolkit configuration controls which tools each agent can access
  • -Agents only see tools from their assigned toolkits during execution
  • -Toolkit assignment managed through the agent editor UI
  • -Prevents agents from accessing irrelevant integration tools
NEWv1.2.0February 25, 2026

Bring Your Own API Key

Use your own Claude, Gemini, or OpenAI API keys for unlimited usage beyond your plan's token allocation. Keys are stored encrypted and used transparently -- you just pay your plan fee plus your own API costs with zero markup.

  • -Support for Claude, Gemini, and OpenAI API keys
  • -Keys encrypted at rest before storage
  • -Automatic fallback to platform keys when user keys are not set
  • -Usage tracking continues to work for analytics regardless of key source
NEWv1.1.0February 22, 2026

Workspace Memories

A persistent memory system that lets your agents remember brand guidelines, preferences, and past decisions. Memories are indexed for fast intelligent semantic search during agent execution.

  • -Agents automatically extract and store important context as memories
  • -Intelligent semantic search for fast and relevant memory retrieval
  • -Memory management page for viewing, editing, and deleting stored memories
  • -Workspace-scoped so all agents in a workspace share the same knowledge base