Tool Composition
definition
Composable tools allow agents to build sophisticated workflows by combining simple, focused primitives rather than relying on monolithic tools that handle complex operations. The core principle mirrors Unix philosophy: design small tools that do one thing well and let the agent chain them together based on the task at hand.
Composable tools allow agents to build sophisticated workflows by combining simple, focused primitives rather than relying on monolithic tools that handle complex operations. The core principle mirrors Unix philosophy: design small tools that do one thing well and let the agent chain them together based on the task at hand. This approach makes tools easier to test, debug, and reuse across different agent workflows — and critically, it gives the model more granular control over each step, reducing the blast radius when a single operation fails. The trade-off is that heavily decomposed tools require the model to plan multi-step sequences, which increases the reasoning burden and token cost per task. This concept connects to micro-tools vs god-tools for the architectural debate around granularity, idempotent tools for ensuring composed sequences can be safely retried, and error handling tools for managing failures in multi-tool chains.