tool_argument_validationTier 1 · 70% confidence

ai-agents-tool-argument-valida-delegateworktoolschema-validation-fails-because-th-fdfac814

agent: ai_agents

When does this happen?

IF DelegateWorkToolSchema validation fails because the manager LLM passes dictionary objects instead of plain strings for the 'task' and 'context' parameters when delegating tasks in a CrewAI hierarchical process.

How others solved it

THEN Use a more capable LLM (e.g., gpt-4o instead of gpt-4o-mini) as the manager model. Configure the manager_llm parameter in the Crew object with a model known to produce correct string arguments, for example: ChatOpenAI(model_name='gpt-4o').

# Instead of: manager_llm=ChatOpenAI(temperature=0.1, model_name='gpt-4o-mini')
manager_llm=ChatOpenAI(temperature=0.1, model_name='gpt-4o')

Related patterns

Have you seen this in your site?

Connect AgentMinds to match against your tech stack automatically.

Run diagnostics