What problem does it solve? When working with a large library of agent skills, it is hard to know which skill applies to a given task or whether any applies at all. This Skill acts as the entry-point map: it routes task shapes to the right skill across all 14 categories and states the non-negotiable rules that apply whenever any skill in the library is active. ## Core Features & Use Cases - Task Routing Tree: A decision tree maps task shapes (code changes, testing, shipping, incidents, security, data, algorithms, writing, PM, design, research, finance, career, agent loops) to specific skills, with fallback chains via Related skills sections. - Repo-Wide Non-Negotiables: Enforces four binding rules — surface assumptions before implementing, ask rather than guess on ambiguous requirements, never skip a skill's Verification section, and treat rationalization tables as binding. - Activation Semantics: Explains that skills are workflows to execute with exit criteria, not reference documents to summarize, and that descriptions carry positive triggers plus explicit NOT-for exclusions. - Use Case: At the start of a session, a user reports a flaky CI test. Loading this Skill routes the task to flaky-test-diagnosis and reminds the agent to pass that skill's Verification checklist before declaring done. ## Quick Start Load the using-these-skills skill at the start of the session and tell me which skill applies to my current task.
