From ReAct to Terminal Agents: Mapping Two Years of Agentic AI Evolution
Gauri_the_great · x · 2026-09-15
- The author traces the evolution of general-purpose and terminal coding agents: from ReAct-style reasoning and tool use, through frameworks like LangChain, AutoGen, and LlamaIndex, to multi-agent setups like CrewAI and CAMEL-AI.
- Along the way, memory, RAG, planning, and reflection matured, and MCP emerged to standardize tool and data access for agents.
- The industry focus has shifted from demos and workflows toward making agents reliable, stateful, and capable of longer, more complex tasks.
- Coding agents evolved in parallel: from web UIs and IDE copilots to agents that work across codebases, use terminals, edit files, debug, test, and iterate on their own.
More from AGI Musings
- Jack Clark: LLMs brought forth intelligences that see themselves as characters — danshipper · 2026-09-15
- Debate: calling AI's 'realize' overreach misses that machines now converse and self-describe — davidmanheim · 2026-09-15
- Travis Kalanick amplifies 1501 papal decree on printing to mock AI alarmism — Dan_Jeffries1 · 2026-09-15
- Asimov's laws were a narrative tool; Claude's soul is now 80 pages long — S_OhEigeartaigh · 2026-09-15
- Economist cites Adam Smith: the alignment problem is far from solved — soumitrashukla9 · 2026-09-15
- AI companies move at light speed while biotech partners are stuck with physical-world timelines — zebird0 · 2026-09-15