The Pragmatic Pivot: Why Anthropic Champions Skills Over Agents

The Pragmatic Pivot: Why Anthropic Champions Skills Over Agents

AI technology is at a critical juncture, grappling with the promise of autonomous agents versus the practicalities of deployment. During a recent panel discussion hosted by Cognition Labs, Barry Zhang, Head of Product and Mahesh Murag, Engineering Lead from Anthropic, provided a compelling argument: the future of reliable AI lies not in building monolithic agents but in developing modular, robust “skills.”

The Limitations of Current LLMs

During the discussion, Zhang and Murag dissected the inherent limitations of current large language models (LLMs) when tasked with complex, multi-step operations. They highlighted that while these models excel at reasoning and planning, their execution reliability diminishes significantly over extended chains of action, leading to unpredictable failures.

Mahesh Murag succinctly articulated this paradigm shift: “We’ve been talking about agents for a long time, but I think what we’re actually building are more like skills.” This distinction is crucial. An agent, in the common conception, is an autonomous entity capable of performing complex, multi-faceted tasks end-to-end. A skill, conversely, is a discrete, well-defined capability designed to perform a specific function reliably.

Why Focus on Skills?

The fundamental challenge, as Barry Zhang elaborated, is that “LLMs are very good at reasoning, but they’re not necessarily good at execution reliability over long chains of actions.” This disconnect between reasoning prowess and execution consistency is the Achilles’ heel of many ambitious agentic projects. An LLM might brilliantly deduce a multi-step plan, but its ability to reliably carry out each step without deviation or hallucination remains a significant hurdle.

This often results in a “hallucination of execution,” where the model believes it has successfully completed a task only for it to fall short in reality. This insight challenges the prevailing narrative and offers a more grounded, engineering-centric approach to AI product development that prioritizes reliability and composability over unfulfilled autonomy.

Attributes of Valuable Skills

Murag emphasized key attributes of a valuable skill: “You want a skill to be robust. You want it to be composable. You want it to be debuggable.” Robustness ensures the skill performs its function consistently, even with minor variations in input.

Composability allows for skills to be combined and recombined in flexible ways, enabling dynamic adaptation to new problems without requiring a complete system overhaul.

Debuggability is paramount for identifying and rectifying failures efficiently, a stark contrast to the opaque, difficult-to-diagnose failures of complex agentic systems. By modularizing AI systems into distinct skills, developers gain finer-grained control and transparency. Failures can be isolated to specific components, making systems easier to audit, monitor, and intervene when necessary.

This human-in-the-loop orchestration of skills offers a more responsible and controllable pathway for deploying AI in sensitive or high-stakes environments. This pragmatic approach acknowledges the current limitations of LLMs while strategically leveraging their strengths, paving the way for more dependable and valuable AI applications.