Chinese Classics Shrimp

Problem

Classic Chinese-text interpretation often suffers from three engineering failures: citations are not checked word-for-word, TV or film plots are mixed into historical sources, and the output reads like a templated AI response. Guoxue Classic Shrimp turns “read the source, draft the interpretation, review, archive” into a repeatable workflow rather than producing one generic summary.

How It Works

It first treats Zhonghua Wenku as the preferred original-text source, looking up chapters from texts such as Dao De Jing, Shiji, and Analects. It then drafts an interpretation with a hook opening, layered development, and a lingering close. The draft must quote the source, explain context and logic, and include about three concrete modern takeaways. It also runs a six-dimension review: source fidelity, factual accuracy, and unsupported-claim check are P0 blockers, while style, logic, and verbosity are lower-priority checks. Approved files are named after the title and chapter, for example 道德经_第二章_解读.md, and synced to the IMA “Guoxue Classics” knowledge base.

Boundaries

It fits chapter, theme, character, and cross-text comparisons, especially when avoiding fabricated citations matters. However, it depends on external source availability; when versions conflict or text is missing, human verification is required. Uncertain points should be marked as open questions instead of being filled in from memory.