Skip to content

Refactor validation responsibilities across validators, manager, and reader #1991

Description

@tuannx

Parent issue: #1834

Problems

  • GoapPathToCompletionValidator logs and continues when a goal has no matching action, so validation can remain falsely valid.
  • DefaultAgentValidationManager aggregates and logs validation errors.
  • AgentMetadataReader logs the same errors again and also applies the deployment policy, causing duplicate logging and mixed responsibilities.

Proposal

  • Validators only return structured ValidationErrors.
  • GoapPathToCompletionValidator returns a dedicated error when a goal has no matching action.
  • DefaultAgentValidationManager remains the composite and single logging boundary.
  • AgentMetadataReader only applies the deployment policy and does not repeat the validation errors.
  • Preserve the current permissive default while making skip-agent-deployment-on-error=true effective for the missing goal-action case.

FQN naming remains in PR #1833 and issue #1990.

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

Labels

No labels
No labels

Type

No type

Projects

No projects

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions