> For the complete documentation index, see [llms.txt](https://0xn3va.gitbook.io/blog/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://0xn3va.gitbook.io/blog/updates.md).

# RSS feed

{% updates format="full" %}
{% update date="2026-03-16" %}

## [Executing AI agent on behalf another user in LangSmith Agent Builder (Resolved)](/blog/2026/langsmith-execute-ai-agent-on-behalf-another-user.md)

This post examines the agent's underlying architecture and the chain of vulnerabilities that led to user impersonation during agent execution.
{% endupdate %}

{% update date="2026-03-16" %}

## [Chaining service key leakage and path confusion in LangSmith (Resolved)](/blog/2026/langsmith-chaining-service-key-leakage-and-path-confusion.md)

This post describes the discovery and exploitation of a vulnerability that allowed unauthorised access across LangSmith agent deployments. In addition, it details research into path normalisation differences between Nginx and GCP Load Balancer.
{% endupdate %}

{% update date="2026-01-12" %}

## [Achieving remote code execution in LangSmith Playground using unsafe template formatting](/blog/2026/langsmith-unsafe-formatting-to-rce.md)

This post details the discovery and exploitation of a vulnerability in LangSmith Playground that allowed arbitrary code execution through unsafe template formatting.
{% endupdate %}
{% endupdates %}