Feat/skill evolution project

[fanqiNO1]

🚀 Core Framework & New Features

• Skill Evolution Workflow: Introduced a foundational framework for signal-driven skill evolution, complete with configuration templates and core pipeline plumbing.
• SearchQA Task Implementation: Added a dedicated SearchQA task suite to serve as the initial implementation and baseline for the evolution workflow.
• Asynchronous Processing Utilities: Implemented robust utility functions, including concurrency-managed orchestration pipelines.

🛠️ Critical Correctness & State Safety Fixes

• Index Alignment on Rollout Failure: Enhanced asynchronous execution handlers to optionally retain None values, preventing downstream index misalignment when intermediate rollouts fail.
• Isolated Configuration Hydration: Replaced shallow-copy mechanisms with explicit deep copies during configuration initialization to prevent unintended cross-agent or global shared state mutations.
• Trajectory Buffer Isolation: Fixed a state-leakage issue where a persistent trajectory buffer caused redundant reflections across iterations, reducing the risk of trajectory overfitting.

⚙️ Refactoring & Technical Debt

• Dataset Lifecycle Adjustments: Corrected the method signature mismatch in BaseDataset.load_data and implemented automatic data reshuffling upon epoch reset to guarantee proper training distribution.
• Configuration Key Alignment: Resolved a critical configuration key mismatch between the runtime parser and schema specifications.
• Lifecycle Management Stubs: Completed dangling TODO blocks regarding macro skill management policies and fallback handling for rejected skill updates.

[gemini-code-assist]

Code Review

This pull request introduces a skill evolution workflow framework, including configuration files, a SearchQA task implementation, and utility functions. The review identifies critical correctness issues regarding index alignment when rollouts fail, which can be addressed by updating gather_with_semaphore to optionally retain None values. Additionally, high-severity issues were raised concerning shared state mutations from shallow-copying configurations, a persistent trajectories buffer that causes redundant reflections and potential overfitting, and incomplete TODO blocks for rejected updates and macro skill management. Other recommendations include fixing a configuration key mismatch, correcting a signature mismatch in BaseDataset.load_data, and reshuffling training data upon epoch reset.

Important

The consumer version of Gemini Code Assist on GitHub is being sunset. Starting June 18, 2026, new organization installations will be blocked, and all code review activity will officially cease on July 17, 2026.
For more details on the timeline and next steps, please review the Help Documentation.

[gemini-code-assist]

Code Review

This pull request introduces a skill evolution workflow framework, including agent configurations, task definitions for SearchQA, and workflow orchestration utilities. The review feedback highlights several critical issues: a bug in gather_with_semaphore that filters out exceptions and causes silent data misalignment during evaluation; a potential crash in _extract_answer when handling non-string or null responses; infinite trajectory accumulation in trajectories_buffer leading to redundant reflections; an incomplete feedback loop due to an unimplemented TODO for rejected updates; a configuration key mismatch for the rejected update buffer size; and an abstract method signature mismatch for load_data in the base dataset class.

Important

The consumer version of Gemini Code Assist on GitHub is being sunset. Starting June 18, 2026, new organization installations will be blocked, and all code review activity will officially cease on July 17, 2026.
For more details on the timeline and next steps, please review the Help Documentation.