How agents work
Each agent runs a sequence of AI-powered steps:- Research — searches Google, reads competitor pages, mines Reddit, pulls Search Console data
- Planning — builds a keyword brief, generates an outline, brainstorms titles
- Execution — writes content section by section, runs QA passes
- Delivery — publishes to your CMS, posts to Slack, or emails you the result
Creating an agent
From the Home chat (recommended for new users) Type your goal in the Home chat. The assistant will ask clarifying questions and launch the workflow automatically. From the Agents page- Click + New Agent in the top right.
- Give the agent a name and optional description.
- Lantern opens the agent builder where you configure tools, goal, and schedule.
The agent builder
The left panel of any agent’s detail page is a conversational interface. Type instructions directly to configure the agent:- “Focus this agent on long-form thought leadership content”
- “Add voice and tone QA to the pipeline”
- “Schedule this to run every Monday at 9am”
Running an agent
Click Run in the top right of the agent detail page to trigger it immediately. A live step list appears in the right panel showing exactly what the AI is doing. What the steps look like:Viewing results
When a run completes, the result appears in the right panel:
For content runs, a rich-text editor opens inline — edit, format, and save directly. If your CMS is connected, a Publish button appears to push content live.
Scheduling
Tell the agent builder in plain English:- “Run every Monday at 8am”
- “Run daily at 6am”
- “Run on the first of every month”
Templates
Pre-built starting points available from the Templates tab:Managing agents
- Search — filter agents by name using the search bar
- Filter by owner — toggle Anyone / Me to see all workspace agents or only your own
- Delete — click
⋯on any agent card → Delete. This permanently removes the agent and all run history