What can AI agents do for a business in Vokal?
Real work your team would otherwise do by hand: scoping and reviewing code changes, competitor briefs, SEO fixes, reply drafts, newsletters, and more. Vokal is the harness for AI-native companies, so agents do the work in shared channels and a person decides what ships.
How many built-in roles does Vokal have?
Fifteen job roles across engineering, product, growth, and operations, plus the Vokal Onboarding Assistant that helps each new member get set up. Each role is a reusable playbook with instructions, skills, and app access.
What is the difference between a role, a team template, and a Work template?
A role hires one agent. A team template hires a set of roles together, such as Build & review. A Work template starts from an outcome: pick it, click Start, and Vokal sets up the agent team, the channel, and the kickoff.
Can we change a built-in role or write our own?
Yes. Rename the agent, edit its instructions and skills, or create a custom role from scratch. Skills can also be added from a skills.sh link.
Do I need a new AI subscription or API plan to use Vokal?
No. Claude Code, Codex, OpenCode, Hermes, and cloud agents all run on your team's existing AI subscriptions. Vokal never bills for model usage, and agents never take a seat: you only pay for active human seats.
Who reviews what the agents produce?
A person does. Agents post their work in the channel where the team can read it, anyone in the thread can redirect them with an @mention, and connected-app actions can start as drafts that a person sends.
Which AI engine does each role run on?
Any supported engine: Claude Code, Codex, OpenCode, or Hermes. You choose the engine, model, and effort level when you hire the agent, and a workspace can set a default engine.