Lizo-RoadTown/claude-skills-marketplace

Decision framework for agent architecture — single vs multi-agent, ReAct vs Plan-and-Execute vs Tree-of-Thoughts, autonomy spectrum, when to introduce an orchestrator.

Liz Osborn's canonical patterns library — ONE home for her reusable agents, skills, and tools. Available in every Claude Code session in every project, so the same pattern means the same thing wherever invoked. v0.1.1 (2026-06-16): adds drift-watcher agent — read-only oversight agent for long multi-repo executions, validated in production twice before promotion.

Designs first-time-user flows using the IDENTITY-TO-HABIT arc — grounded in BJ Fogg's Tiny Habits and Fogg Behavior Model, Nir Eyal's Hook Model, and James Clear's identity-based habits.

Repo-specific schema guard for SDE_Extraction, hook-enforced: a PostToolUse hook runs the repo's scripts/check_schema.py after edits to the extraction surface and blocks a broken schema (the class of bug that crash-looped the worker). Bundles the recorded-transformation discipline skill. Complements loom-discipline (generic drift + architecture snapshots); no-ops in repos that don't have the SDE schema.