Geo Agents Introduction
Updated
The Agents Landing Page serves as a centralized workspace for discovering available GEO agents, launching agent workflows, reviewing AI-generated recommendations, and tracking ongoing agent activities. It enables you to create, manage, and revisit agent-generated content from a single location within LLM Insights.
Scenarios for Using Agents
Discover and launch GEO agents: Access all available GEO agents from a single location and quickly start workflows for content generation, publishing, engagement, planning, or backlink monitoring.
Act on LLM Insights recommendations: Convert visibility opportunities identified in LLM Insights into actionable outputs such as blogs, FAQs, social content, campaign briefs, and engagement responses.
Save and continue work later: Save agent outputs as drafts, allowing you to pause work and return later without losing progress.
Track agent activity and ownership: Monitor the status, creator, and progress of agent runs through the Record Manager, ensuring visibility into ongoing and completed work.
Manage and revisit previous agent runs: Access historical agent records to review generated outputs, continue existing workflows, or reuse previously created content and recommendations.
Centralize AI-powered workflow management: Use a unified workspace to generate content, review recommendations, manage drafts, and track agent activity without switching between multiple tools or screens.
Accessing the Agents Landing Page
Navigate to the Agents tab within LLM Insights to access available GEO agents and previously created agent records.
Explore Available GEO Agents
The landing page displays a collection of GEO agent cards. Each card includes:
Agent icon
Agent name
Short description
Configure button to launch the workflow.
Selecting Configure opens the selected agent in create mode.
The 3 agents are:
Blog Content Recommendation Agent: The Blog Content Recommendation Agent identifies content opportunities based on insights and citations from your LLM Insights report. It recommends high-potential blog topics and content ideas that can help improve your brand's AI search visibility and address uncovered content gaps.
Output: Long-form blogs and content assets.
Content Optimisation Suggestions Agent: The Content Optimisation Suggestions Agent analyzes your existing web content against insights from your LLM Insights report and recommends targeted improvements. These recommendations can include content structure enhancements, stronger headings, improved phrasing, evidence-based claims, and answer-ready content snippets to increase AI visibility and citation potential.
Output: Scheduled posts optimized for LLM visibility.
Reddit Content Recommendation Agent: The Reddit Content Recommendation Agent identifies relevant, high-value Reddit conversations associated with opportunities surfaced in your LLM Insights report. It recommends contextual engagement responses that help your brand participate in discussions, increase visibility, and engage with audiences on topics relevant to your business.
Output: Scheduled posts optimized for LLM visibility.
Create an Agent Run
The steps are listed below:
Navigate to the Agents tab.
Select the desired GEO agent.
Select Configure.
Review the available recommendations.
Choose a recommendation.
Review the generated output associated with that recommendation.
Note:
You can open AI-generated content from GEO agents in a dedicated full-screen editor, giving you more space to review, edit, and refine long-form outputs such as blogs and FAQs. This enhancement improves readability and editing efficiency while allowing you to save drafts and continue working without losing changes.
Access to this feature is controlled by a dynamic property (DP: LLM_INSIGHTS_ENABLED). To enable this feature in your environment, reach out to your Success Manager. Alternatively, you can submit a request at tickets@sprinklr.com.
Review Recommendations and Generated Output
When an agent is launched, a full-screen workspace opens with two sections:
Recommendations Panel
The left panel displays a list of available recommendations. Each recommendation includes:
Title
Description
Relevant source or target links, when available.
You can select one recommendation at a time.
Once a recommendation is selected, the system generates content related to that recommendation. The right panel displays the generated output, which can be reviewed and edited before taking action. Generated content is labeled as Sprinklr AI generated content.
Save and Manage Agent Workflows
Save generated content as a Draft to continue working on it later. Enter a draft name and save it to create a persistent record.
Use the Record Manager to track agent runs, including the agent name, status, creation timestamp, and ownership details.
Reopen previously saved agent records to review outputs or continue existing workflows.
Cancel a workflow at any time if you decide not to proceed. Discarding a workflow closes the experience without creating an agent record.
All agent activities are managed within a full-screen workspace where recommendations and generated outputs are displayed side by side for easy review.
Available actions and generated outputs may vary depending on the selected GEO agent.