Frequently asked questions
Everything you need to know about Lobsterpod AI agents.
Getting started
- What is Lobsterpod?
- Lobsterpod is a managed AI agents platform. You describe the job — lead follow-up, content drafting, business admin — and Lobsterpod deploys an AI agent that handles it for you. The agent runs in an isolated environment, communicates via Slack or WhatsApp, and operates within hard budget caps you set.
- How quickly can I get an agent running?
- Under 60 seconds. Sign up, choose an agent type (Sales, Ghostwriter, Secretary, or Tradesman), describe your business and what you need, set a budget cap, and deploy. There is no configuration, no code, and no infrastructure to manage.
- Do I need technical knowledge to use Lobsterpod?
- No. Lobsterpod is built for business owners, not developers. You describe the job in plain language. The platform handles everything else — model selection, prompt engineering, environment isolation, and monitoring.
- What kind of businesses use Lobsterpod?
- Primarily service businesses — tradespeople, consultants, real estate agents, coaches, and small agencies — who need to automate customer communication, lead follow-up, and admin without hiring staff or managing software.
Agents and use cases
- What counts as an agent?
- Each agent is a single AI worker with one job. A Ghostwriter drafts content. A Sales agent qualifies leads. A Secretary handles admin. A Tradesman answers WhatsApp enquiries. Running two agents counts as two agents against your plan.
- Can an agent do multiple jobs?
- Each agent is optimised for one job — this is by design. A Sales agent that also tries to write content does both poorly. If you need two types of automation, use two agents. The Team plan supports three agents.
- Can I customise what my agent does?
- Yes. During setup you tell the agent your services, tone, qualification criteria, notification preferences, and how much autonomy it has. You can adjust these at any time. The Ghostwriter agent also learns from your content over time.
- Which messaging platforms do agents work with?
- Agents communicate via Slack (for you) and can respond via WhatsApp, email, or website chat (for your customers). The Tradesman agent is specifically built around WhatsApp.
Credits and pricing
- What are usage credits?
- Usage credits are what your agents spend when they work — reading data, writing content, making decisions. Each plan includes a monthly credit allocation. When credits are used up, agents pause until the next billing cycle (or until you upgrade).
- How do credits work across multiple agents?
- Credits are shared across all agents on your plan. If you are on Team with three agents, they all draw from the same credit pool. Heavier tasks (research, long-form content) use more credits than lighter ones (short messages, qualification questions).
- What happens when my agents run out of credits?
- All agents on the plan pause immediately. You get a notification. You can wait for the next billing cycle (when credits reset) or upgrade to a higher plan for more credits.
- Is there a free trial?
- Not yet — but every plan comes with a hard budget cap, so you will never spend more than the listed price. You can start on the Lite plan at $59/mo and cancel anytime.
- Can I switch plans?
- Yes. You can upgrade or downgrade at any time. Changes take effect on your next billing cycle. Unused credits do not roll over.
Budget caps and cost control
- How do budget caps work?
- You set a hard spending limit per agent. When the cap is reached, the agent stops making API calls — no exceptions, no overages. You stay in control and never wake up to a surprise bill.
- Can agents run up unexpected costs?
- No. The hard budget cap is a hard stop, not a soft warning. If your agent hits the cap, it pauses. You are never billed more than the price of your plan.
- What is model routing and does it affect my costs?
- On managed plans, Lobsterpod picks the best model for each task. Research tasks may route to Perplexity. Deep reasoning tasks go to a powerful model. Routine messages use the cheapest model available. You get the best results at the lowest cost without configuring anything.
Security and data
- How is my data kept secure?
- Every agent runs in its own isolated environment with zero access to your personal machine or other agents. Inside that environment, agents follow strict rules — they can only use skills that have been vetted and approved, and only run the types of commands they are allowed to run. There is no data sharing between agents.
- Does Lobsterpod have access to my customer data?
- Lobsterpod processes data through the agent to fulfill its job (e.g. reading an enquiry email to qualify a lead). This data is processed in an isolated environment and is not used to train models. See our privacy policy for full details.
- Is Lobsterpod GDPR compliant?
- Lobsterpod is designed with GDPR compliance in mind. Customer data is processed only to fulfill the agent's job and is not retained beyond what is necessary. Full details are in our privacy policy.