Skip to content
Search Sign in List your company

Silmaril

Silmaril builds runtime security for AI systems, aimed at prompt injection: the attack where the instructions come in through the data an agent was asked to read.

No reviews yet San Francisco, United States Small Founded 2026

About Silmaril

Silmaril is runtime security for AI. Its subject is prompt injection, which is the attack that has no equivalent in ordinary software: an agent reads an email, a web page or a document, and something written in that content is treated as an instruction rather than as data. The defence sits between the agent and what it reads, and it retrains itself as new attacks appear rather than waiting for a rule to be written. That last part is the argument for it: a filter that only knows yesterday's attacks is a filter with a shelf life, and agents are being handed real permissions faster than any static list can keep up with. The company was founded in 2026 by Aum Upadhyay and Eduardo Velasco, went through Y Combinator's Spring 2026 batch, and is based in San Francisco.

Client reviews

No reviews yet

Be the first to review Silmaril.

Frequently asked questions

What is prompt injection?
An attack where hostile instructions are hidden in content an AI system was asked to read - an email, a web page, a support ticket - and the system acts on them as if you had typed them. It matters more as agents are given real permissions, because the instruction and the data arrive through the same channel.
What does Silmaril do about it?
It sits at runtime between the agent and what it reads, catches injection attempts, and retrains itself as new attacks appear rather than waiting for somebody to write a rule.
When was Silmaril founded, and where is it based?
Founded in 2026 by Aum Upadhyay and Eduardo Velasco, based in San Francisco, and part of Y Combinator's Spring 2026 batch.