From b1186d331eeb790525094d58f7293d2c6ec2b8d9 Mon Sep 17 00:00:00 2001 From: Mika Lanir Date: Mon, 2 Nov 2026 20:27:36 +0200 Subject: [PATCH 1/2] Add a usage responsibility notice to the AI agents overview. Readers need to see that they configure and audit third-party AI clients, and that least privilege and human approval are recommended. Co-authored-by: Cursor --- ai-agents.md | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/ai-agents.md b/ai-agents.md index c54b612..e00d00c 100644 --- a/ai-agents.md +++ b/ai-agents.md @@ -4,6 +4,12 @@ Extend your AI agents with custom, deterministic actions powered by JFrog Worker Explore ready-to-use integration samples and practical use cases to turn agent decisions into reliable actions across your DevSecOps workflows. +> **Note** +> +> **Usage Responsibility Notice**: When integrating the JFrog Platform with third-party AI clients or agents, you are responsible for the configuration, management, and auditing of those integrations in accordance with your organization's policies. The behavior of third-party AI agents is governed by the terms and practices of their respective providers, not by JFrog. +> +> We recommend configuring your AI client to operate under least-privilege principles, granting access only to the resources required for your specific workflows. Additionally, we strongly recommend enabling human-in-the-loop approval in your AI client settings, so that explicit user confirmation is required before any tool execution that creates, updates, or deletes resources in your JFrog Platform. + Each sample folder includes `worker.ts`, `manifest.json`, and a README with the secrets, properties, and setup steps. ## After Download From bf972d5b40a49228d130e0aa98c0a5bcefc83093 Mon Sep 17 00:00:00 2001 From: Mika Lanir <110965216+MikaLJF@users.noreply.github.com> Date: Mon, 2 Nov 2026 22:16:24 +0200 Subject: [PATCH 2/2] Update usage responsibility notice in ai-agents.md Removed recommendation for least-privilege principles and human-in-the-loop approval. --- ai-agents.md | 2 -- 1 file changed, 2 deletions(-) diff --git a/ai-agents.md b/ai-agents.md index e00d00c..70a7846 100644 --- a/ai-agents.md +++ b/ai-agents.md @@ -7,8 +7,6 @@ Explore ready-to-use integration samples and practical use cases to turn agent d > **Note** > > **Usage Responsibility Notice**: When integrating the JFrog Platform with third-party AI clients or agents, you are responsible for the configuration, management, and auditing of those integrations in accordance with your organization's policies. The behavior of third-party AI agents is governed by the terms and practices of their respective providers, not by JFrog. -> -> We recommend configuring your AI client to operate under least-privilege principles, granting access only to the resources required for your specific workflows. Additionally, we strongly recommend enabling human-in-the-loop approval in your AI client settings, so that explicit user confirmation is required before any tool execution that creates, updates, or deletes resources in your JFrog Platform. Each sample folder includes `worker.ts`, `manifest.json`, and a README with the secrets, properties, and setup steps.