Preprint
Nov 2026
AgentScript-Eval: Benchmarking LLMs and Agents for Code Generation in an Enterprise DSL
AgentScript-Eval is introduced, an execution-graded benchmark of 100 real tasks for AgentScript, the declarative DSL that Salesforce uses in Agentforce to build conversational agents.
Dipin Khati, Shubham Mehrotra, Bin Bi et al.
· 0 citations