It is based on logic programming and the belief–desire–intention software model (BDI) architecture for (cognitive) autonomous agents.
In 1996, Anand Rao created a logic-based agent programming language based on the BDI architecture and named it AgentSpeak(L).
In its original conception, AgentSpeak was an abstract agent programming language aimed to help the understanding of the relation between practical implementations of the BDI architecture such as procedural reasoning system (PRS)[2] and the formalisation of the ideas behind the BDI architecture using modal logics.
[4] In recent years, the language has been revived both in research and development.
AgentSpeak(L) is also the most prominent means for developing the solution for the Multi-Agent Programming Contest from the beginning of the competition up to the recent edition.