What’s New in Agent.ai: Product Updates, Features & Improvements

There's always something new to explore in Agent.ai.

And now you'll know just where to find it.

Every time we ship a feature, refine a workflow, or smooth out a rough edge, we'll add it here. Some updates will be big. Others will just make your day that much easier. All of them are designed to give you smarter agents, help you discover better agents, and make sure you get more value from the agents you use.

Read on for what's new.

🎬 YouTube Creator Team—From Idea to Published Content in One Flow

Creating consistent video content usually means juggling research, scripting, packaging, and promotion across multiple tools. The new YouTube Creator Team connects all of it into one continuous workflow, so you can go from “what should I make?” to fully packaged and promoted content without starting over at each step.

What’s new:

• Idea Researcher generates personalized video ideas with titles, angles, and real-time trend signals
• Script Writer produces production-ready scripts with structure, pacing, and built-in cues
• Thumbnail Generator creates titles, descriptions, tags, and thumbnail images in one run
• YouTube to Social Content turns one video into platform-ready posts for LinkedIn, X, Instagram, email, and blog

Each step carries context forward, so you stay in flow instead of re-entering information. The result: faster content cycles and more consistent output.

🧠 Knowledge Agent APIs—Bring Conversational Agents Into Your Apps

Knowledge agents are no longer limited to the Agent.ai interface. With the new API support, you can now integrate them directly into your own products and workflows while preserving full conversation context.

What’s new:

• Invoke knowledge agents via API with the same responses as the web experience
• Maintain stateful conversations using run IDs to continue chats over time
• Retrieve past conversations with titles and history for easy resume
• Discover knowledge agents programmatically with clear input/output definitions

This opens the door to embedding rich, context-aware conversations anywhere you need them.

🛠️ Agent Builder—More Accurate Workflows, Better Outputs

Building agents just got more reliable. Updates to the builder improve how workflows are generated, validated, and structured—so you get outputs you can trust without extra cleanup.

What’s new:

• Agent Assistant now only suggests actions you actually have access to
• Validation recognizes dynamic tools and integrations instead of flagging them as errors
• Structured outputs let you define JSON schemas for consistent, predictable results
• Anthropic skills enable direct generation of files like PDFs, PowerPoints, Word docs, and spreadsheets

Together, these changes reduce guesswork and make it easier to build agents that work correctly the first time.

🖼️ Image Workflows—Save, Reuse, and Create Without Friction

Working across image-generating agents is now faster and more connected. You can save your best results, reuse them instantly, and move between workflows without extra steps.

What’s new:

• Save generated headshots and avatars to a shared collection with one click
• Access all saved images from a unified “Saved” tab across agents
• Send saved images directly into new workflows without downloading or re-uploading
• Choose from saved images when uploading instead of starting from scratch

Plus, the HeyGen Avatar Video agent now includes more avatar options with visual previews, making it easier to pick the right fit before generating your video.

These updates make it simpler to iterate, reuse, and build on what you’ve already created.

🔍 Market Research Team—From One-Off Research to Continuous Intelligence

Understanding your market shouldn’t require hours of manual digging. The new Market Research Team gives you a coordinated set of agents that turn scattered research tasks into a structured, always-on intelligence workflow.

What’s new:

• A full suite of research agents covering industry, competitors, trends, companies, and prospects
• Structured industry briefs with market size, segments, and key players
• Ongoing competitor monitoring with on-demand or recurring updates
• Trend tracking with digestible summaries of emerging shifts
• Company and prospect research to help you walk into meetings prepared

Each agent works on its own, but together they create a system that keeps you informed without constant effort.

⚙️ Agent APIs—Deeper Visibility, More Control

It’s now easier to understand and integrate agents without trial and error. New API capabilities let you inspect how an agent works before running it, so you can build more reliably on top of Agent.ai.

What’s new:

• A new describe_agent endpoint to view an agent’s inputs, outputs, and metadata
• Support for both agent IDs and human-readable slugs
• Visibility into key details like usage, authors, and supported fields
• Consistent authentication with existing /v1/action/* endpoints

This makes it simpler to choose the right agent and integrate it with confidence.

💬 Knowledge Agents via API—Bring Conversations Into Your Product

The same conversational agents you use in the Agent.ai interface are now fully accessible via API. That means you can embed rich, stateful AI conversations directly into your own workflows and applications.

What’s new:

• Programmatic access to knowledge (chat) agents with invoke_agent
• Full responses returned in a single call, using the same logic as the web experience
• Stateful conversations using run_id to maintain context across interactions
• A new endpoint to list and resume past conversations

You can now move beyond single prompts and build continuous, context-aware interactions powered by your agents.

🔎 Agent Marketplace—Find the Right Agent Faster

The agent directory has been rebuilt into a true marketplace experience, making it much easier to discover the right agents for your workflow. With powerful search, deeper filtering, and curated shortcuts, you can quickly narrow down thousands of agents to the ones that best match what you need.

What's new:

• A redesigned Agent Marketplace at agent.ai/agents with fast, full-text search
• 9 sidebar filters including Category, Integrations, Pricing, Author, Rating, and more
• Ability to select multiple filters at once to refine results more precisely
• Preset filter shortcuts like Hot Right Now, New & Noteworthy, and Sales Prospecting Agents
• A filter for cloneable agents so you can quickly find agents you can duplicate and customize
• Shareable filtered URLs, making it easy to send curated agent lists to teammates

The result: faster discovery, better matches, and less time digging through the directory.

🔗 HubSpot Integration—Easier Setup for HubSpot-Powered Agents

If you're using agents that rely on HubSpot, getting connected just got simpler. HubSpot now has its own dedicated section in Settings so you can quickly link your portal or start a free account without hunting through integrations menus.

What's new:

• A new HubSpot tab in Settings that puts HubSpot connection front and center
• A direct linkable setup page, allowing agents and onboarding flows to send you straight to HubSpot configuration
• A "Get HubSpot Free" option for users who don't yet have a HubSpot account
• A cleaner Integrations page that now links directly to the new HubSpot setup location

This reduces friction when setting up HubSpot-powered agents and makes it faster to get them running.

🤝 Meeting Intelligence Team—An End-to-End Meeting Workflow

Great meetings depend on preparation, context, and timely follow-up. The new Meeting Intelligence Team connects the entire workflow—from company research to post-meeting action items—so insights carry forward automatically and nothing slips through the cracks.

What's new:

• Company Research turns a company name into a structured intelligence brief with funding data, tech stack, hiring signals, and recent news in under a minute.
• Prospect Research compiles public data, recent activity, and company context into a clear contact profile.
• Meeting Prep automatically generates a briefing on attendees, past interactions, and key discussion points before your call.
• Meeting Follow-Up drafts action items, follow-up emails, and CRM-ready notes after the meeting.

The result: less time preparing and documenting, and more time focusing on the conversation itself.

🎯 Sales Prospecting Team—From Lead Discovery to First Outreach

Prospecting often breaks down between steps: finding leads, researching them, prioritizing opportunities, and writing outreach. The new Sales Prospecting Team connects those stages into a single workflow so research and insights flow forward—helping you quickly decide who to pursue and why.

What's new:

• Prospect Finder lets you describe your ideal customer in plain language and returns a ranked list of potential leads
• Prospect Researcher builds structured profiles using public data, recent activity, and company context
• Lead Qualifier scores prospects against your ICP and recommends pursue, nurture, or pass
• Outreach Drafter generates personalized email and LinkedIn messages grounded in the research

Instead of starting from scratch each time, you get a clear path from identifying prospects to sending a relevant first message.

🧩 Meeting Prep Agent v3—From Research to Strategy

This upgrade makes the agent sales-strategy-aware, not just research-aware— so your brief reflects what you sell and how you win.

What's new:

• Scan-first structure: Briefs are organized by call flow—Person → Company → Situation → Segues → Questions → Next Step Plan. No long paragraphs. Just what you need in the order you'll use it.
• Seller Profile integration: Teach the agent what you sell, who you sell to, your value props, proof points, competitors, and sales process. The brief now aligns to your actual motion.
• Evidence-backed segues: Get 3–5 ready-to-use lines that connect something specific about them to something real about you.
• Stage-aware next steps: Specific asks, A/B closes, quick wins, and logistics tailored to meeting type.
• Confidence-gated attendee insights: "Show me you know me" personalization—useful, grounded, and appropriately cautious.

If you haven't opened Meeting Prep in a while, it's worth another look. The difference isn't cosmetic. It's strategic.

🔐 Google Docs-Style Agent Sharing

Agent sharing just got an major upgrade.

We've rebuilt agent sharing from the ground up. The old "Sharing & Visibility" dropdown in the agent experience is gone. In its place you'll find a dedicated Share tab that works just like sharing works in Google Docs.

For Builders

You can now:

• Share agents by email by assigning one of three sets of permissions by role:
• Viewer - run only
• Commenter - run + view configuration
• Editor - run + view + edit configuration
• See exactly who has access, complete with real names and avatars
• Change roles or remove access instantly
• Choose Anyone with the link or Restricted access
• Submit agents to the public directory directly from the Share tab
• Control agent-to-agent invocation with a simple toggle
• Changes save instantly — no more "Publish Changes" for sharing settings

For Agent Users

Visiting a restricted agent now looks and feels more familiar. You'll see a clean "You need access" page where you can request Viewer, Commenter, or Editor access — with an optional note.

Owners get an email, approve or deny from the Share tab, and once approved, the page updates automatically. No refresh. No guessing.

In Your Agent List

• Shared agents now appear in your agent list with clear role badges
• Clone is available to Editors and above
• Delete is reserved for owners
• Status now shows Restricted (instead of "Restricted Users")

It's cleaner, faster, and much harder to share something with the wrong permissions.

✉️ Custom Instructions

Too often working with AI means repeating yourself — instructions, background information, other kinds of context — over and over again. Now you can teach your agent your preferences just once. And move on.

There's a new Email Agent tab in Settings where you can define how the agent should respond: tone, format, language, structure — whatever matters to you.

No instructions set? Nothing changes.

Less re-explaining. More consistency.

🔌 OpenClaw Actions

Workflows can now talk directly to OpenClaw.

OpenClaw actions are live in the Action Library, which means builders can connect workflows to a user's personal OpenClaw instance without duct tape or manual config.

Three new actions:

Get Details
Drop it in first. It automatically resolves the user's instance credentials — no IPs or tokens required.

Chat Completion
Send messages through the standard OpenAI-compatible API. Supports multi-turn conversations, session persistence, and streaming.

Tools Invoke
Call any tool available on the user's OpenClaw Gateway directly from a workflow step.

You can find all three actions in the new OpenClaw category.

🧠 Knowledge Agent Evaluations

Now you can turn regular conversations into real answers.

Knowledge agent builders can now review user conversations and curate few-shot examples directly from the builder.

There's a new Evaluations tab under Knowledge & Eval that shows message pairs from real interactions.

Builders can now:

• Toggle messages on or off as examples
• Edit responses to create ideal answers
• Automatically inject selected examples into the agent's system prompt at runtime

By the numbers:

• Up to 100 examples per agent
• 5,000-character budget for prompt injection
• Polls for new messages every 10 seconds
• Zero impact if no examples are configured

It's agent training that's grounded in reality, not guesswork.

⚡ Event Triggers UX Overhaul

Triggers now live right where they belong.

You can browse, connect, and configure event triggers directly from the Edit Trigger sidebar. No more hopping into the Actions Library to wire things up.

What's new:

• Third-party action vendors show their connection status upfront
• One-click Connect handles OAuth inline
• Required fields are validated before creation, with clear, human error messages
• Active triggers display in the builder as:
  Trigger Agent On: Email Sent (with a clear icon)

🤖 Agent Builder Assistant — Major Upgrade

The Agent Assistant now generates higher-quality agents, faster, and with far fewer surprises.

For builders:

• Actions are generated with guaranteed valid structure
• Supports complex workflows with 2x larger output capacity (32k tokens) using the latest Claude Opus 4.6
• Self-correcting generation—the assistant validates and refines before delivering
• Live progress updates during generation
• Generated actions save immediately and persist across refreshes
• Fully editable after applying

On reliability: We fixed several edge cases around action editing, template variable validation, and draft persistence. Added 71 automated tests covering the full generation-to-validation pipeline.

Smarter generation. Stronger guardrails. Better agents for you.