Agents & workflows15 guidesUpdated
AI agents and workflow automation guides
Build agents from templates, a canvas or code, and test them free.
Every Agents & workflows guide
- Start hereBuild your first agentBuild your first AI agent in Suede AI: describe the job, review every step, test it free with a dry run, then launch it to work around the clock.
- Guided, Studio and CodeDescribe it in Guided, wire it on the Studio canvas, or read it as TypeScript in Code. One saved agent, three views, and you switch any time.
- TemplatesSkip the blank page. Open one of more than 100 ready-made AI agent templates or a CRM workflow starter, fully wired and ready for a free test.
- Triggers and schedulesRun agents on a schedule, from a webhook, or the moment something happens in your CRM. Set a trigger in Suede AI Agent Studio and the work starts itself.
- Test free with dry runsRun a full dry-run test of any agent for free. Watch every step stream its result, catch wiring mistakes, and go live only when the output is right.
- Versions, Test and LiveSave every agent change as a numbered version, see exactly what changed, roll back in a click, and promote the exact version you tested to Live.
- Runs and costsEvery agent run in Suede AI leaves a receipt: status, trigger, timing, each step's result and its cost. Filter, fix failures and rerun in one click.
- Connect appsConnect Slack, Google Search Console, Analytics, GitHub, your CRM and any API to Suede AI. Keys are encrypted on arrival and never shown again.
- Website to agentPaste your URL and Suede AI reads your public pages and drafts an agent that answers from them. Review every detail, then launch it in minutes.
- Self-driving companiesSet a destination and objectives, pick an autonomy tier, and Suede AI's AI employees plan, run and check the work nightly, escalating what needs you.
- Knowledge and ResourcesGround your agents in your own documents and data. Use Knowledge Search inside a run, or build a reviewed, versioned Resource Pack in Resource Foundry.
- Wait and branch stepsBuild follow-ups that pause, check and adapt. Use Wait and Branch steps to time your messages and send each contact down the right path.
- Lead qualification agentBuild an AI agent that scores every inbound lead 1 to 10 on fit and intent, explains why, names the next step and files it in your CRM before you call.
- Write agent promptsWrite AI agent prompts that return the same clean output every run: role, rules, exact JSON shape and your own facts, tested free before going live.
- Human approval and escalationKeep a person in charge of the moments that matter: approval gates, draft-then-approve replies, and instant hand-offs to your team when an agent needs you.
Questions buyers ask
Do I need to know how to code to build an AI agent?
No. Guided builds the agent from a plain-English description and a few questions, and Studio lets you adjust it by dragging and connecting steps. Code is there for developers who want TypeScript, and nobody else ever has to open it.
Is Agent Studio a no-code AI agent builder?
Yes. Guided and Studio build complete agents without a line of code, and the Code view is there for teams who also want TypeScript. All three work on the same agent.
Are Agent Studio templates free to use?
Opening, editing and dry-run testing any template costs nothing. Once it runs live, the agent draws from your prepaid credit like any other agent; rates are on the pricing page.
What time zone do schedules use?
Schedules run in UTC so they never shift with daylight saving. The inspector shows the equivalent in your local time right under the picker, so you always know when the agent will run.
Is testing an AI agent in Agent Studio really free?
Yes. Every dry run is free, with no limit on how many you run. Live runs, where AI steps call the model and messages actually send, use prepaid credit.
Does saving a version publish my agent?
No. Saving freezes a checkpoint and nothing else. Your agent reaches customers only when you promote a tested version to Live.
Is the Runs page like GoHighLevel's workflow execution logs?
It covers the same job and adds cost. Every run gets its own receipt with each step's output, status, error text and estimated cost, plus one-click rerun and file downloads.
Does Agent Studio work with apps that have no built-in tile?
Yes. Connect any service with a bearer token, API key header, username and password, or custom headers, then call it from the HTTP Request node in any agent.
More of the help center
- Get startedYour first win in minutes: setup, modes, plans and the tour.
- CRM & salesContacts, deals, inbox, calendars, forms and reports in one CRM.
- AI employees & phoneAn AI receptionist, salesperson and support agent on every line.
- SEO & AI searchRank on Google and get quoted in AI answers.
- Agency & clientsWin clients, onboard them fast and prove the work every week.
- Marketing & reputationCampaigns, social, reviews, sites, courses and communities.
- Payments & invoicingInvoices, estimates, payment links, products and subscriptions.
- DevelopersAPI, MCP, SDK, CLI, webhooks and agent publishing.
- Help & accountSign-in, billing, security, troubleshooting and support.