langchain_agent_intermediate_steps
Number of intermediate steps (thoughts/actions) before agent finishesDimensions:None
Available on:
OpenTelemetry (1)
Interface Metrics (1)
Dimensions:None
Related Insights (3)
Runaway Token Consumption Cost Spikecritical
Recursive chains, agent loops, or unbounded context windows can generate thousands of tokens in seconds, causing unexpected cost explosions (e.g., $12k-$30k bills).
▸
Agent Infinite Loop Detectioncritical
Agents can enter infinite tool-calling loops when tool selection logic fails or tools return results that trigger repeated invocations, consuming tokens and delaying responses.
▸
Agent Intermediate Steps Explosionwarning
langchain_agent_intermediate_steps counts reasoning/tool-use iterations. Unbounded growth indicates agents spinning on complex tasks, inefficient tool selection, or poor stopping criteria, driving up latency and cost.
▸