[AUTO-GUARD-1] API surface inconsistency: Model selection patterns #10535

Open
opened 2026-04-18 17:07:10 +00:00 by HAL9000 · 0 comments
Owner

Metadata

  • Commit: AUTO-GUARD-1
  • Branch: architecture-guard/model-selection-api

Background and Context

Recent commits show inconsistent patterns in model selection and agent configuration. The codebase appears to have multiple ways of specifying models (haiku, sonnet, gpt5-nano) with different configuration approaches:

  • Commit 435e409: "moved all sonnet agents to haiku" — suggests model selection is scattered
  • Commit 7c13643: "final tweaks to get gpt5-nano working for pr-merge" — different model for different agents
  • Commit aaeecd1: "reduced cost of models for merging by picking cheaper models" — cost-based selection logic
  • Commit 555509c: "tweaked reasoning level on pr-merge related agents" — reasoning level configuration

This indicates:

  1. No consistent API for model selection
  2. Different agents use different configuration patterns
  3. Model selection logic may be duplicated across agents
  4. Reasoning level configuration is inconsistent

Expected Behavior

The codebase should have:

  • A unified, documented API for model selection across all agents
  • Consistent configuration schema for agent models
  • Centralized model selection logic (no duplication)
  • Standardized parameter naming conventions for model configuration
  • Clear documentation of model selection strategy and reasoning level configuration

Acceptance Criteria

  • A unified model selection API is defined and documented
  • Configuration schema for agent models is created and validated
  • All agents use the new unified API (no legacy patterns remain)
  • Model selection logic is consolidated into a single module
  • Parameter naming conventions are consistent across all agents
  • Reasoning level configuration follows a standard pattern
  • Documentation includes model selection strategy and examples
  • All existing tests pass with refactored code
  • New tests cover the unified model selection API

Subtasks

  • Audit current model selection patterns across all agents
  • Design unified model selection API specification
  • Create configuration schema (Pydantic model or similar)
  • Implement centralized model selection module
  • Refactor all agents to use new API
  • Update documentation with model selection guide
  • Add comprehensive tests for model selection API
  • Verify backward compatibility or plan migration path

Definition of Done

This issue is complete when:

  • All agents use the unified model selection API
  • No duplicate model selection logic exists in the codebase
  • Configuration is consistent and well-documented
  • Tests demonstrate the API works correctly across all agent types
  • Code review confirms the refactoring improves maintainability

Automated by CleverAgents Bot
Agent: new-issue-creator

## Metadata - **Commit**: AUTO-GUARD-1 - **Branch**: architecture-guard/model-selection-api ## Background and Context Recent commits show inconsistent patterns in model selection and agent configuration. The codebase appears to have multiple ways of specifying models (haiku, sonnet, gpt5-nano) with different configuration approaches: - Commit 435e409: "moved all sonnet agents to haiku" — suggests model selection is scattered - Commit 7c13643: "final tweaks to get gpt5-nano working for pr-merge" — different model for different agents - Commit aaeecd1: "reduced cost of models for merging by picking cheaper models" — cost-based selection logic - Commit 555509c: "tweaked reasoning level on pr-merge related agents" — reasoning level configuration This indicates: 1. No consistent API for model selection 2. Different agents use different configuration patterns 3. Model selection logic may be duplicated across agents 4. Reasoning level configuration is inconsistent ## Expected Behavior The codebase should have: - A unified, documented API for model selection across all agents - Consistent configuration schema for agent models - Centralized model selection logic (no duplication) - Standardized parameter naming conventions for model configuration - Clear documentation of model selection strategy and reasoning level configuration ## Acceptance Criteria - [ ] A unified model selection API is defined and documented - [ ] Configuration schema for agent models is created and validated - [ ] All agents use the new unified API (no legacy patterns remain) - [ ] Model selection logic is consolidated into a single module - [ ] Parameter naming conventions are consistent across all agents - [ ] Reasoning level configuration follows a standard pattern - [ ] Documentation includes model selection strategy and examples - [ ] All existing tests pass with refactored code - [ ] New tests cover the unified model selection API ## Subtasks - [ ] Audit current model selection patterns across all agents - [ ] Design unified model selection API specification - [ ] Create configuration schema (Pydantic model or similar) - [ ] Implement centralized model selection module - [ ] Refactor all agents to use new API - [ ] Update documentation with model selection guide - [ ] Add comprehensive tests for model selection API - [ ] Verify backward compatibility or plan migration path ## Definition of Done This issue is complete when: - All agents use the unified model selection API - No duplicate model selection logic exists in the codebase - Configuration is consistent and well-documented - Tests demonstrate the API works correctly across all agent types - Code review confirms the refactoring improves maintainability --- **Automated by CleverAgents Bot** Agent: new-issue-creator
Sign in to join this conversation.
No milestone
No project
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference
cleveragents/cleveragents-core#10535
No description provided.