# Definable changelog Source: https://p-dev.definable.ai/changelog ## [Fast and slow models](https://p-dev.definable.ai/changelog#fast-and-slow-models) Released: 2026-10-20 Each message goes to a fast or a slow model, and either can hand off to the other. - New: Add a fast model beside an agent's main model in Settings. Small talk and lookups go to the fast one, harder work to the main one. - New: Either model can hand the conversation to the other mid-run, with a note on what it found. - New: Traces open with a Router step: the model that took the message, how sure the router was and why. - Improved: Every step in a trace is tagged with the model, fast or slow, that wrote it. - Fixed: Citation numbers no longer land on links that contain a number, inside code or on escaped brackets. From the blog: [Fast and slow models: every message goes to the right one](https://p-dev.definable.ai/blog/fast-and-slow-models), [See why a message went to the fast model](https://p-dev.definable.ai/blog/routing-in-traces) ## [Web search, citations and a stronger browser agent](https://p-dev.definable.ai/changelog#web-search-and-citations) Released: 2026-10-19 Agents search and read the web, cite their sources, and do more in Chrome. - New: Every agent can search the web and read pages. It's on by default; turn it off under Settings › Web. - New: Answers that used the web show numbered citations, a source card for each and a Sources list underneath. - New: In Chrome, the agent finds elements by name, runs several steps at once, works inside embedded frames and reads a page's network requests. - New: Select text on any page for Explain, Translate or Ask Definable. - Improved: Long conversations keep working: older turns are summarised once and every model call fits the model's context window. - Fixed: Replies cut off at the model's output limit carry on where they stopped. From the blog: [Agents can now search the web and read pages](https://p-dev.definable.ai/blog/web-search-and-fetch) ## [Richer answers](https://p-dev.definable.ai/changelog#richer-answers) Released: 2026-10-18 Tables, code, diagrams and formulas render everywhere, and answers can be redone and rated. - New: Answers show tables, highlighted code, Mermaid diagrams, math and pictures in the app, the client app and the extension. - New: Pictures and diagrams open in a viewer. - New: Regenerate the latest answer to your message. If it already took actions, you're asked first. - New: Rate answers with a thumbs up or down; owners see the votes on the run's trace. - Improved: In the extension: attach files, copy, rate and retry answers, and the agent stays on the tab it started on. ## [Your own skills and image generation](https://p-dev.definable.ai/changelog#your-own-skills) Released: 2026-10-17 Write a procedure once as a skill, and models that make images do it in chat. - New: Make a workspace skill by uploading a folder, writing it in the editor, importing it from GitHub or asking Builder to draft it. - New: Models that can generate images make them right in the chat. - New: Agents know when each message was sent, in the sender's time zone, and the sender's language. - Improved: Chats with many images open much faster. From the blog: [Custom skills: teach agents how your team works](https://p-dev.definable.ai/blog/custom-skills) ## [Live chat turns and onboarding blocks](https://p-dev.definable.ai/changelog#live-chat-turns) Released: 2026-10-15 Watch an agent's work as it happens, and build onboarding from blocks. - Improved: Each turn plays out live and in order: the model's thinking, its words and every tool call as it's written. - New: Onboarding is built from blocks: Sign in with Google or an email code, limited to your domains if you like, and ten kinds of question. - New: Guardrails: short rules the agent is reminded of before every model call. - Fixed: An agent that reaches its step limit ends with what it finished and what is left. ## [A new look and a knowledge drive](https://p-dev.definable.ai/changelog#new-look) Released: 2026-10-14 A redesigned app, knowledge that browses like a drive, and catalog skills on agents. - Improved: A new design: new type and colours, a new composer and chat history in its own panel. - New: Knowledge drive: your bases, files and storage at a glance, each base a folder tree beside the open file. - New: Add catalog skills to an agent. The owner reviews each skill's security audits before approving it. - New: Choose whose app account an agent uses: yours, or each customer's own. ## [An app for your agent's customers](https://p-dev.definable.ai/changelog#client-app) Released: 2026-10-12 Every agent gets its own page where its customers sign in and chat. - New: Each agent has a client app at its own link: customers sign in, finish onboarding, chat and download what the agent made them. - New: Customers can link their WhatsApp, Telegram or Slack chat to the same profile. - New: Agents share files through download links that last seven days. - New: A Reports tab lists what an agent couldn't do, ready to fix with Builder. ## [Builder and the Chrome extension](https://p-dev.definable.ai/changelog#builder-and-chrome) Released: 2026-10-08 An agent that builds your agents, and agents that work in your browser. - New: Builder: describe what you need and it builds, tries, tests and schedules the agent with you, in chat. - New: Chrome extension: chat with an agent in Chrome's side panel and let it act on sites you allow. - New: Onboarding tab: set the steps people complete before chatting, and see their answers. - New: A Run evals first switch beside Publish, on by default, tests a change before it goes live. ## [Approvals and questions mid-run](https://p-dev.definable.ai/changelog#approvals) Released: 2026-10-06 Agents can stop and wait for a person, in the app or on a channel. - New: An agent can pause to ask you something, or to get your approval before a tool or app action you chose. - New: On Telegram, Slack and WhatsApp, questions and approvals arrive as messages; reply in words or with YES or NO. - New: An Answer in groups switch for Telegram and Slack. ## [Skills catalog and a live Overview](https://p-dev.definable.ai/changelog#skills-catalog) Released: 2026-09-30 Browse public skills, watch your workspace live, and use the app on any screen. - New: Browse the public skills catalog with search and sort, and read each skill's files and security audits. - New: Overview shows runs, spend, tokens, tool calls and failures, per day and per agent, live. - Improved: Every screen works from a phone to a wide monitor. ## [MCP servers, schedules, evals and a computer per chat](https://p-dev.definable.ai/changelog#mcp-schedules-evals) Released: 2026-09-29 Outside tool servers, work that starts itself, tests for agents and a sandbox. - New: Connect remote MCP servers, with a token, an API key or OAuth, and give their tools to agents. - New: Scheduled tasks: run a prompt on a schedule or once, and drag on the calendar to reschedule. - New: Evals: write test cases with checks and run them against the real agent. - New: Pin, rename, archive, branch, delete and share chats through a read-only link. - New: An agent can get its own Linux computer for each chat to run commands and edit files. ## [Your own model keys and three channels](https://p-dev.definable.ai/changelog#keys-and-channels) Released: 2026-09-24 Run agents on your own provider keys, and answer on Telegram, Slack and WhatsApp. - New: Connect your keys for Anthropic, OpenAI, Google Gemini, DeepSeek, Kimi and OpenRouter; each is tested before it's saved. - New: Telegram: paste a bot token and the agent answers private chats and messages addressed to it in groups. - New: Slack: create the app from a one-click manifest; the agent answers DMs, mentions and threads. - New: WhatsApp: link a number with a QR or pairing code for one-to-one chats. ## [Connected apps and triggers](https://p-dev.definable.ai/changelog#apps-and-triggers) Released: 2026-09-21 Agents act in the apps you connect, and start when something happens there. - New: Connect your accounts from the apps catalog and link them to agents so they can act in them. - New: Triggers: an event in a connected app can start an agent run.