Terms & Conditions

AgentCC provides analytics and controls for AI agents — token-usage and cost tracking, waste detection (loops, prompt bloat, redundant calls, expensive models, call spikes), per-agent budgets, and an auto-stop kill switch — via a dashboard and an SDK that wraps your existing LLM client.

You are responsible for the activity under your account and for keeping your credentials and API keys secure. Notify us promptly of any unauthorized use.

Do not misuse the Service: no unlawful activity, no attempts to disrupt or reverse-engineer the platform, no probing other users' data, and no use that violates the terms of the LLM providers you connect.

The SDK transmits usage metadata only — token counts, cost, latency, and a one-way prompt fingerprint. It never transmits your prompts, completions, or LLM provider API keys. You are responsible for the keys you issue within AgentCC and how you deploy the SDK.

The Service works alongside third parties you choose (e.g. OpenAI). Your use of those services is governed by their own terms, and their charges are your responsibility.

The Service is provided "as is" and "as available." We do not guarantee uninterrupted or error-free operation, and the auto-stop / kill features are best-effort safeguards, not guarantees against overspend.

To the maximum extent permitted by law, AgentCC is not liable for indirect, incidental, or consequential damages, or for LLM provider costs incurred while using the Service. The Service does not constitute financial advice.

You may stop using the Service at any time. We may suspend or terminate access for breach of these Terms.

We may update these Terms; material changes will be reflected by the "Last updated" date above. Continued use after changes means you accept them.

Questions about these Terms? Contact the AgentCC team. See also our Privacy Policy.