
CustomGPT
Your Tars AI agent connects to CustomGPT to list projects, pull conversation histories, manage licenses, and check usage limits. Teams get instant visibility into their CustomGPT deployment without logging into another dashboard or memorizing API calls.




Your agent handles CustomGPT project administration, from listing active agents to reviewing chat logs and managing license keys, all within a single conversation thread.
CustomGPT
Real scenarios where teams use AI agents to monitor their CustomGPT deployments, audit conversations, and keep their AI projects running smoothly.
A customer success manager asks 'Show me the last conversation from our support bot.' The AI agent identifies the CustomGPT project, retrieves the most recent session ID, and pulls the complete message thread. The manager reviews the exchange right in the chat, spots a hallucination issue, and flags it for retraining. No CustomGPT login needed, no session IDs to look up manually.
Your operations team asks 'Are we close to our CustomGPT limits this month?' The AI agent calls the usage limits endpoint and returns current consumption versus caps for queries, agents, and word counts. If you are at 85% of your query limit with ten days remaining, the team knows to upgrade or throttle usage before customers experience downtime.
After a product pivot, several CustomGPT demo bots are no longer needed. Your team messages 'List all our CustomGPT projects.' The agent returns the full inventory. They identify five inactive bots and confirm deletion one by one. The agent removes each project through the API. Subscription costs drop, and the workspace stays organized.

CustomGPT
FAQs
The agent calls CustomGPT's List Conversation Messages endpoint using the project ID and session ID. It returns the full exchange between the end user and the CustomGPT bot, including timestamps and message content. Your team can review these conversations for quality assurance without logging into the CustomGPT dashboard or searching through session logs.
Yes. The agent calls the Get Usage Limits endpoint which returns your current agent count, word usage, and query consumption against your plan caps. It provides a clear picture of where you stand relative to your limits, helping you decide whether to upgrade your plan or optimize bot usage before hitting overages.
Tars needs a CustomGPT.ai API Key generated from your profile's API tab. This key authenticates all requests. The key provides access to project management, conversation retrieval, license operations, and usage monitoring. You can create a dedicated key for Tars and revoke it at any time from your CustomGPT profile settings.
No. When the agent retrieves conversation messages, the data is fetched live from CustomGPT's API and displayed in the current conversation only. Tars does not maintain a separate copy of your CustomGPT chat histories. Once the conversation ends, the fetched data is not persisted on Tars servers.
The agent can delete projects via the Delete Project endpoint, but deletion is permanent. There is no undo. The agent will confirm the project ID before executing, and you can configure approval workflows so deletions require explicit confirmation. We recommend backing up project configurations before removing agents.
The CustomGPT dashboard requires manual login, navigation, and clicking through project settings. Tars lets your team ask questions like 'list my projects' or 'show usage limits' and get instant answers. It is especially useful for teams managing multiple CustomGPT agents who want a unified, conversational interface across all their tools.
Yes. The agent can list all licenses tied to a project, retrieve details for a specific license by ID, update license names, and delete licenses. This is useful for teams distributing CustomGPT access across departments or clients, where license administration needs to happen quickly without dashboard switching.
The agent will receive an authentication error from CustomGPT's API and notify your team that the connection is broken. No data is exposed or corrupted. Simply generate a new API key in your CustomGPT profile and update the credentials in your Tars dashboard to restore the integration.
Don't limit your AI Agent to basic conversations. Watch how to configure and add powerful tools making your agent smarter and more functional.

Privacy & Security
At Tars, we take privacy and security very seriously. We are compliant with GDPR, ISO, SOC 2, and HIPAA.