ToolSimulator: scalable tool testing for AI agents
AWS just shipped ToolSimulator: LLM-powered testing for AI agents without the PII risk. Catch bugs at scale before production.
Why it matters
As enterprises deploy AI agents with external tool integrations, safe testing at scale becomes critical. AWS's ToolSimulator solves the real problem: validating multi-turn agent workflows without exposing PII or triggering live API calls—shipped today in Strands Evals SDK.
The key facts
12 to knowProduct: ToolSimulator (LLM-powered tool simulation framework)
Released: April 20, 2026 (immediate availability)
Part of: Strands Evals SDK (AWS)
Solves: Safe agent testing without PII exposure, edge case validation, multi-turn workflow support
Use case: Production-ready AI agents with external tool dependencies
Key benefit: Replaces risky live API calls and brittle static mocks
ToolSimulator available as part of Strands Evals SDK
LLM-powered tool simulation framework eliminates need for live API calls during testing
Reduces PII exposure risk and unintended action triggers
Supports multi-turn workflow validation
Enables edge case testing at scale
AWS product announcement
Go to the source
AWS Machine Learning Blogaws.amazon.com
Publisher excerpt: You can use ToolSimulator, an LLM-powered tool simulation framework within Strands Evals, to thoroughly and safely test AI agents that rely on external tools, at scale. Instead of risking live API calls that expose personally identifiable information (PII), trigger unintended actions, or settling…