dsh-zhouli
Run the following command in DeepSeek Harness:
dsh plugin install SCP-CN-1059/dsh-zhouli
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install SCP-CN-1059/dsh-zhouli in your terminal to install; source code: https://github.com/SCP-CN-1059/dsh-zhouli
About this plugin
When an LLM recites classical Chinese from memory, there is no way to verify whether a passage actually exists. dsh-zhouli turns the six-book, 49,384-character Zhou Li into a coordinate-addressed corpus: every paragraph carries a book.official index (e.g. 1.1 for the Grand Tutor), so any citation can be traced back to the source line, eliminating fabricated or misattributed quotes.
The tool offers seven actions: full-text search, passage retrieval by coordinate, official-role lookup with headcounts, chapter listing, paragraph dump, collation between the base text and Zheng Xuan commentary (567 recorded variants), and corpus statistics. A bundled Zhou Li Mode preset wires the six-official framework, the eight self-review methods, and the year-end six-count audit into a persistent agent persona, requiring coordinate-anchored citations at every step from kickoff through close.
Built for classical-studies researchers, digital-humanities pipelines, and agent teams that need provable, verifiable citations from pre-Qin canonical texts. The core requirement is simple: every line the agent quotes from Zhou Li must have a real, checkable coordinate behind it.
Screenshots
Use Cases
- Agent cites Zhou Li passages with book.official coordinates so every quote is traceable to the source line.
- Structured search, official-role lookup, and textual collation across the full 49,384-character corpus in digital-humanities pipelines.
- Apply the Zhou Li Mode preset to discipline an agent's reasoning via the six-official framework and year-end audit.
Best For
- Pre-Qin and classical-textual scholars
- Digital humanities and computational humanities teams
- Agent-workflow developers requiring verifiable citations
Related Plugins
A method pack that makes AI coding agents plan against your real baseline, prove completion with fresh evidence, and reduce reworks and unsafe changes.
Turns the DeepSeek Harness session into a captain that builds a durable sub-agent team, splits goals into dependency-aware tasks, and coordinates work via direct messages and a live Web UI.
Gives coding agents design judgment, letting Claude Code, Cursor, and 70+ agents generate and iterate high-quality UI, presentations, and graphics right from the terminal.
Run the Pi ecosystem's plugins on DeepSeek Harness, unmodified, via a compatibility layer that implements Pi's public extension ABI on DSH's native services.