September 2026

Every connected account now carries its own approval rules, and they apply to everything that uses it: your agents, your coordinator, and any MCP client. Open Settings → Credentials and click the shield on a connection to reach its Tool permissions page.

A locked action runs nothing until you say so. The call waits, the exact arguments it would send are shown to you, and you approve or decline it.

• The Use without approval row sets everything at once. Read-only lets reading actions run and locks the rest, All lets every action run, and None locks everything, including actions the service adds later. Search the list to lock one action on its own.
• Requests arrive under Needs you on the Dashboard, where you can read the arguments and decide without leaving the page.
• Approving admits one call with exactly those arguments, and it has 30 minutes to run. Different arguments, a second attempt, or a change to the rules each need a fresh approval. An untouched request expires after 24 hours.
• Agents and MCP clients can add restrictions but never remove one. Unlocking happens only on the permissions page, signed in as the connection's owner.
• Operations you allowlist on a tool provider node now carry a lock in the node settings, so you can see and change what needs approval without leaving the workflow.

The Twilio node gained two call operations, so a number you already own can carry a voice agent.

• On Incoming Call is a trigger. Pick the Phone Number that should ring, write a Greeting, and NoClick points that number's voice webhook at your workflow. The agent wired to the trigger answers and holds the conversation live.
• Place an Agent Call is a tool an agent calls itself. Give it a From Number, a To number, and a Goal in plain words. It returns straight away while the call runs.
• A finished call comes back as a message in the agent's chat, with the transcript and how long the call lasted. An outbound call reports back to the agent that placed it, in the conversation it was placed from, so a workflow no longer needs an On Incoming Call node to hear about a call it made.
• Both operations need an Account SID and Auth Token credential, because an API key cannot verify the webhooks Twilio signs. One number rings one agent, and a second workflow that tries to claim it is told which workflow holds it.

Every account now has a coordinator: one assistant that sees the whole account rather than a single workflow. It reads your workflows and runs, hands builds and fixes to the AI builder, passes work to agents you already have, and connects accounts for you.

• It watches failed runs. When a workflow fails, the coordinator looks into it: it asks the builder to diagnose and fix, pauses a workflow that is clearly finished with, or tells you when only you can fix it. A run that failed for want of credits never wakes it, because the credits alert already covers that.
• Memories is a new card on the Dashboard. It holds what the coordinator has learned about you and your work, as entries you can read, edit, write yourself, and delete. Each one is filed as About you, Preferences, Project context, or Reference.
• Preferences live there too, rather than in a settings page. Tell the coordinator to stop triaging a workflow, or which channel to reach you on, and it keeps that as a memory.
• It reaches you where you last wrote to it, on WhatsApp or by email. It has its own inbound address on noclick.app, so you can mail it and it answers on the same thread.
• It makes images and video on request. Video generation is available on the Plus and Pro plans.
• Your agents can message it as well, so an agent that hits something it cannot fix has somewhere to send it.

A voice note sent to an agent on WhatsApp, Telegram, or Discord used to arrive as a link the agent could not listen to, so the best it could do was ask the sender to type it out. Audio that arrives with a message is now transcribed, and the transcript becomes part of the turn the agent reads, so it answers what was actually said. A document that arrives the same way has its text pulled out at no cost, and photos and videos keep their link.

• This covers WhatsApp, Telegram, Discord, attachments on an inbound email, and files you upload in the agent chat. Telegram attachments are also saved to the workflow's resources as the message arrives, so a voice note plays in the run and chat views instead of showing an ID you cannot open.
• Files shared in a Slack message reach the agent too, up to three per message and 25 MB each. Slack keeps its file links private to the workspace, so NoClick saves a copy to the workflow's resources and gives the agent a link it can fetch.
• The transcript appears under the audio player in the agent's chat and in the run view, so you can check what the agent heard.
• Transcription is on by default and can be turned off per agent: Transcribe Voice Messages under Advanced in the node's settings, or Voice transcription under Capabilities in the agent chat's Settings. With it off, the agent is told that a voice message arrived and can ask the sender for text.
• Transcription costs credits like other AI usage. Audio is transcribed up to 20 MB and 10 minutes, at most three audio files per message. Anything larger, longer, or in a format the transcriber does not accept becomes a note telling the agent to ask the sender for text instead. A Test Run transcribes nothing, because the media it stages is made up.

The Discord node has two new triggers, On Channel Message and On Bot Mention, and a message reaches your workflow within seconds of being posted. Connect Discord with the Discord Bot Install credential so NoClick's bot is in your server, then leave Channel empty to listen to the whole server or pick one channel. Ignore bot messages is on by default, so posts from other bots and webhooks do not wake the agent. Each channel and thread is its own conversation.

• A Discord message in run results reads as a message: who wrote it, the channel and server by name, mentions as names, and attachments by filename. A slash command reads as its invocation.
• A Test Run on a Discord-triggered agent ships a written Channel question situation, so the rehearsal looks like a real Discord message instead of a generic sample.
• An On Slash Command trigger fires only for the server its connection was installed into. A connection made with your own bot token still covers every server that bot is in.
• A Discord connection reads as disconnected once the bot is removed from its server, in the credential picker, in workflow validation, and to an agent asking about its own workflow, with reconnecting named as the fix.