What problem does it solve? When building a knowledge wiki, common background concepts get redefined repeatedly across pages. This Skill seeds terminal foundation pages under wiki/foundations/ so future ingestion can link to shared definitions without duplicating them. ## Core Features & Use Cases - Wikipedia-backed seeding: Fetches public encyclopedia content via a fetcher script, handling disambiguation, missing pages, and network errors with interactive fallbacks. - Duplicate detection and refresh: Checks whether a foundation already exists and offers skip, refresh, or abort options while preserving created dates and user-edited sections. - Alias management and linting: Proposes unique aliases for each concept and runs lint.mjs --fix to keep wiki/index.md and structural checks clean. - Use Case: Before ingesting project sources about machine learning, run this Skill to create a foundation page for "Reinforcement Learning from Human Feedback" with aliases like "RLHF", so later source pages link to it instead of redefining the term. ## Quick Start Ask the agent to prefill a foundation page for a background topic, for example: create a foundation page for the topic "Reinforcement Learning from Human Feedback".
