Workflows are the automations your agents run. A trigger starts one, steps check conditions and use skills, and actions send the email, add the contact to a list or notify Slack. Routines run an agent's instruction on the days and time you pick. Your team builds and tests the workflows, you decide which steps wait for a person to approve, and every run shows the result of each step.
No credit card. 500 free execution actions to start.
By signing up you agree to graph8's Terms and Privacy Policy.
You pay execution credits for what the agents run, not per seat.
Programmable revenue is the graph8 platform plus the APIs, MCP server and SDKs to run it and build on it. Workflows are where you program it inside the app. You set the trigger, the steps and the approvals, and the agents follow them.
Starts the run with the contact from the form
Researches the contact's company
Yes: draft outreach. No: add to a nurture list
Uses the research from the step before
The run pauses until someone approves or rejects
Enrolls the contact once the email is approved
A test run uses sample data and shows the result of each step without running the real actions.
Stop, continue, retry with a delay, or route to a fallback step. You pick the rule for each action.
Each feature works in the app. Where supported, your team can also use it through the MCP server or the CLI.
Start a workflow when a form is submitted, a new intent signal or website visitor comes in, a deal changes stage, a webhook fires, or someone runs it by hand.
Drag steps onto a visual canvas, write a simple step list, or edit the JSON. Switching modes converts the workflow. Or start from a template such as Lead Qualification or Meeting Preparation.
A skill is a reusable AI action with a fixed prompt, set inputs and an output format. Start from a template such as Company Research or Contact Dossier, then attach it to an agent or add it to a workflow as a step. Other actions call an API, run a script or run a whole workflow.
Steps can send an email, add a contact to a list or a sequence, assign the contact to a rep round robin, enrich a contact, create a deal or a quote, send a LinkedIn message, SMS or WhatsApp, and notify Slack.
An If/Else step branches on the data, for example to skip outreach when the lead score is too low. A Loop step repeats the work for each item in a list.
A Human Approval step pauses the run until someone approves or rejects it. Put one before anything that cannot be undone, such as sending an email or enrolling a contact in a sequence.
A Handoff step passes the work, with its context, to another agent or to a person on your team.
For each action step, you set what happens when it fails: stop, continue, retry with a delay and backoff, or route to a fallback step. Every run shows running, completed, failed or paused, with the result and any error for each step.
Pick an agent, write the instruction, choose the days, the time and the timezone, and pick the Work app conversation that receives the result.
A twin of each rep, trained on their own mail, calls and LinkedIn.
Explore In AgentsForecasts pipeline, flags at-risk deals, drafts the weekly review.
Explore In AgentsBuilds territory lists, drafts sequences, routes replies.
Explore In AgentsWatches usage, flags churn, drafts renewal briefs.
Explore In AgentsProposals, redlines, pricing comparables, case studies.
Explore In AgentsDrafts pages for search and tracks citations in AI answers.
Explore In AgentsArticles, ads, landing pages, enablement, on brand.
Explore In AgentsDynamic web and email content per account.
Explore In AgentsAutopilot, Approve or Off, reviewer queues and audit logs.
ExploreExecution credits send, call, and run agents. Contact credits reveal and enrich the people you target. Three ways in.
A free entry point to the full platform. No credit card.
Unlimited contact data, in the tools you already use.
Unlimited users and data. Every app, one bill.
The $99 Team plan is the full platform: everything below, for the whole org. MCP Unlimited and Pay as you go are lighter entry points with usage and seat limits.
Execution credits send, call, and run agents. Contact credits reveal and enrich the people you target.
Sign-up takes about ten seconds with Google or Microsoft. Then an agent sets up the account with you. Start from a template, test it with sample data, and turn it on when you are ready. Free to start, no card.
One email. Two weekly sends: the Tenbound research report with the invite to the Tuesday community call, and the graph8 changelog with the invite to the weekly demo. Written by the Institute and the builders. Unsubscribe any time.