0.1.0

Added

• Initial local Agent Learning System.
• record-agent-learning skill for writing reusable session learnings into the
  Obsidian inbox.
• consolidate-agent-learnings skill for processing inbox and reviewed notes,
  promoting reusable lessons, archiving processed notes, and writing reports.
• scripts/agent_learning.py helper for store initialization, note capture,
  consolidation preparation, note finalization, report writing, notification,
  and privacy scanning.
• Explicit record-agent-learning hook mode for detecting local review skills
  and adding an idempotent learning-capture hook only when requested.
• Repository source synchronization for hook mode when installed skills are
  copied locally instead of symlinked from their source repository.
• Filesystem-backed learning store configuration where AGENT_LEARNING_DIR
  names the base directory and AGENT_LEARNING_STORE_NAME names the generated
  AI Agent Learnings folder.
• Installer for config generation, learning store initialization, copy-based
  skill installation, Codex skill mirrors, automation records, and local
  validation.
• Codex automation prompt files for midnight consolidation, noon
  consolidation, and morning review email.
• Unit tests covering note capture, review decisions, finalization, and
  outside-store rejection.
• README documentation with Mermaid diagrams for install, note lifecycle, and
  hooked review-skill, and scheduled automation flows.
• Repository-local agent guidance in AGENTS.md.