-
Notifications
You must be signed in to change notification settings - Fork 0
Agents Finding
GT AI OS Release edited this page Jun 11, 2026
·
3 revisions
Gen 3: This is a legacy Gen 2 article. For current GT AI OS 3.0 guidance, see gen3/agents.
This guide explains how to discover and browse available agents in GT AI OS.
Click Agents in the sidebar to view the agent gallery.
The gallery displays agents as cards showing:
- Agent name
- Brief description
- Visibility indicator (personal, team, organization)
- Quick action buttons
Use filters to narrow down agents:
- Category: Filter by agent category (e.g., General Purpose, Technical)
- Tags: Filter by tags assigned to agents
- Creator: Filter by who created the agent
Sort agents by:
- Recently Used
- Global Usage
- Most Used
- Name
- Date Created
Use the search bar to find agents by name or description keywords.
Each agent card shows:
- Name: The agent's identifier
- Description: What the agent does
- Owner: Who created the agent
- Visibility: Who can access it
From the agent card you can:
- Click the card: Start a conversation with this agent
- Edit (pencil icon): Edit the agent configuration (if you have permission)
- Delete (trash icon): Delete the agent (if you have permission)
To view an agent's full configuration:
- Find the agent in the gallery
- Click the Edit (pencil) icon on the agent card
- The configuration panel shows:
- Basic settings (name, description)
- System prompt
- Model configuration
- Attached datasets
- Visibility settings
- Easy buttons
Think about what you need help with:
- Research and writing: Look for general-purpose agents
- Code and technical work: Find technical agents
- Domain-specific questions: Check for domain expert agents
- Creative work: Use creative-focused agents
- New to the platform? Start with general-purpose agents
- Specific domain? Look for domain expert agents
- Team projects? Check team-shared agents first
- Custom needs? Consider creating your own
- Favorite frequently used agents by starring them for quick access
- Check agent descriptions carefully before selecting
- Try multiple agents to find the best fit for your task
- Ask colleagues which agents they recommend