Skip to content

Enhancements to AI Scriptless - #45

Merged
diego-ferrand merged 8 commits into
ARG-AMBIGUITYfrom
AI-SCRIPTLESS-ALIGNMENT
Sep 2, 2026
Merged

Enhancements to AI Scriptless#45
diego-ferrand merged 8 commits into
ARG-AMBIGUITYfrom
AI-SCRIPTLESS-ALIGNMENT

Conversation

@3dgiordano

@3dgiordano 3dgiordano commented Aug 19, 2026

Copy link
Copy Markdown
Collaborator

This pull request introduces significant enhancements to the AI Scriptless formatter and its related models and tests. The main focus is on improving the representation, validation, and editing of script step arguments and parameters, providing more detailed metadata, and supporting richer contract-based command definitions. Several utility functions and data models have been added or refactored to support these improvements.

Formatter and Model Enhancements:

  • Added new models: ScriptStepArgument, ScriptStepParameter, and ScriptStepDetail in models/ai_scriptless.py to represent detailed step argument and parameter metadata, including validation rules, allowed values, data sources, and display labels.
  • Implemented format_step_detail and supporting utility functions in formatters/ai_scriptless.py to provide a comprehensive view of a script step, including its arguments, unset parameters, and editing notes.
  • Improved argument value formatting and display logic, including better handling of multivalued parameters and secured values.
  • Enhanced display names for loops and conditions to better reflect variable-driven and labeled constructs.

Command Contracts and Validation:

  • Updated the offline test fixtures to use a more comprehensive command contract model, supporting richer parameter and command metadata for validation and editing. The fixture was renamed from declare_command_parameters to declare_commands and now handles full command contracts, not just parameter lists.

Documentation and Policy Clarifications:

  • Updated command selection policy and argument binding documentation to clarify the use of DataTable bindings and validation against declared types, ranges, and allowed values. [1] [2]

Test and Utility Adjustments:

  • Adjusted test utility functions and test cases to align with new argument and contract models, including changes to how IfStatement and group elements are constructed for tests. [1] [2]
  • Added format_step_detail to the test imports to enable direct testing of the new detailed formatting.

These changes improve the reliability, clarity, and editability of scriptless test steps, making it easier to inspect, validate, and modify script steps programmatically or through the UI.

@3dgiordano 3dgiordano changed the title Ai scriptless alignment Enhancements to the AI Scriptless formatter and its related models Aug 19, 2026
@3dgiordano 3dgiordano changed the title Enhancements to the AI Scriptless formatter and its related models Enhancements to the AI Scriptless Aug 20, 2026
@3dgiordano 3dgiordano changed the title Enhancements to the AI Scriptless Enhancements to AI Scriptless Aug 20, 2026
@3dgiordano
3dgiordano marked this pull request as ready for review August 20, 2026 00:12
@diego-ferrand
diego-ferrand merged commit 5561fe0 into ARG-AMBIGUITY Sep 2, 2026
5 checks passed
@diego-ferrand
diego-ferrand deleted the AI-SCRIPTLESS-ALIGNMENT branch September 2, 2026 18:27
diego-ferrand added a commit that referenced this pull request Sep 2, 2026
* Better handling of ambiguity in arguments

* Update tests

* Update uv.lock version

* Better handling of ambiguity in arguments

* Accept arguments as a cmd_arguments alias and cover normalize_action_args.

* Enhancements to AI Scriptless (#45)

* Support dual store and data type

* Improve status, condition and error policy

* Support condition statement and error policy

* Variable, command contract and validations

* Update hint notes

* Contract and others improvements

* Contract/spec and improvements

---------

Co-authored-by: diego-ferrand <diego.ferrand@abstracta.com.uy>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants