HAL9000
8a5567f5d8
Implemented a new context policy configuration loader and integrated plan execution context assembly for ACMS. Key additions include: - New module: src/cleveragents/acms/context_policy_loader.py - ContextPolicyConfigurationLoader class for loading YAML/TOML configurations - Data models: PolicyScope and ContextPolicyConfig dataclasses - ViewPolicyConfiguration for per-view policy management - Schema validation to ensure policy configurations adhere to expected structure and constraints - Supports loading configurations from both files and strings, with robust error reporting - New module: src/cleveragents/acms/plan_execution_integration.py - ACMSContextAssembler for assembling runtime context based on policy-driven decisions - PlanExecutionACMSIntegration to connect with the plan execution engine - Flexible policy loading from files or strings, allowing runtime configurability - BDD tests - features/acms_context_policy_loader.feature (20 scenarios) validating loader behavior and policy scoping - features/acms_plan_execution_integration.feature (8 scenarios) validating end-to-end plan-context integration - features/steps/acms_context_policy_loader_steps.py (step definitions) - features/steps/acms_plan_execution_integration_steps.py (step definitions) - Tests cover YAML/TOML parsing, validation errors, per-view policy application, and plan integration flows - Updated exports - Updated src/cleveragents/acms/__init__.py to export the two new modules, enabling easier imports and usage ISSUES CLOSED: #9584
63 lines
2.7 KiB
Gherkin
63 lines
2.7 KiB
Gherkin
Feature: Plan Execution ACMS Integration
|
|
As a developer
|
|
I want the plan execution engine to use ACMS-assembled context for LLM calls
|
|
So that LLM calls receive properly assembled context instead of raw file dumps
|
|
|
|
Background:
|
|
Given I have a plan execution ACMS integration
|
|
|
|
Scenario: Prepare LLM context without policy configuration
|
|
Given I have no policy configuration
|
|
When I prepare LLM context with raw context data
|
|
Then the LLM context should be the same as the raw context
|
|
|
|
Scenario: Prepare LLM context with policy configuration
|
|
Given I have a policy configuration with 1 policy
|
|
When I prepare LLM context with raw context data
|
|
Then the LLM context should be assembled using ACMS policies
|
|
|
|
Scenario: Load policy configuration from file
|
|
Given I have a policy configuration file
|
|
When I load the policy configuration from the file
|
|
Then the integration should have the policy configuration loaded
|
|
|
|
Scenario: Load policy configuration from YAML string
|
|
Given I have a YAML policy configuration string
|
|
When I load the policy configuration from the YAML string
|
|
Then the integration should have the policy configuration loaded
|
|
|
|
Scenario: Load policy configuration from TOML string
|
|
Given I have a TOML policy configuration string
|
|
When I load the policy configuration from the TOML string
|
|
Then the integration should have the policy configuration loaded
|
|
|
|
Scenario: End-to-end: Plan execution with ACMS context
|
|
Given I have a plan execution ACMS integration
|
|
And I have a policy configuration with scope rules
|
|
And I have raw context data from file analysis
|
|
When I prepare LLM context for plan execution
|
|
Then the LLM context should include applied policies
|
|
And the LLM context should include assembled data
|
|
And the LLM context should have the correct view name
|
|
|
|
Scenario: ACMS context assembly respects priority weights
|
|
Given I have a policy configuration with multiple policies
|
|
And policy1 has priority_weight 1.0
|
|
And policy2 has priority_weight 2.0
|
|
When I prepare LLM context
|
|
Then policy2 should be applied before policy1 in the assembled context
|
|
|
|
Scenario: ACMS context assembly applies budget overrides
|
|
Given I have a policy configuration with 1 policy
|
|
And the policy has budget_override 500
|
|
When I prepare LLM context
|
|
Then the assembled context should have budget 500
|
|
|
|
Scenario: ACMS context assembly filters by scope
|
|
Given I have a policy configuration with scope rules
|
|
And the scope rule is file_type equals python
|
|
When I prepare LLM context with file_type "python"
|
|
Then the policy should be applied
|
|
When I prepare LLM context with file_type "javascript"
|
|
Then the policy should not be applied
|