caveman
claudeTalk like caveman. Cut filler. Keep technical details explicit.
22 skills · 1 command · 3 agents · hooks
Extensible research workflows coordinated across local files, pluggable executors, and deterministic validation.
Plugin yanshu-workbench in panzhzh/yanshu-workbench. Commands read from the repo's manifest — 1 month ago.
This plugin runs code on your machine — it starts local MCP servers. The exact commands are listed below; read them before installing.
citation-auditReview and strengthen citations in an academic manuscript with YanShu by verifying claim-source fit, identifying missing citations, validating BibTeX, and safely repairing citation issues when requested. Use when the user asks YanShu or 研术台 to audit references, check citation correctness or coverage
experimental-plottingCreate reproducible, publication-ready experimental plots with YanShu from authentic result files, metric definitions, statistical protocols, and manuscript context. Use when the user asks YanShu or 研术台 to plot experimental results, produce a paper chart, visualize ablations or sensitivity, or repla
idea-discoveryFind and verify research ideas with YanShu from recent literature, a field or problem seed, dataset conditions, and resource limits, then save aligned Chinese and English Markdown reports. Use when the user asks YanShu or 研术台 to discover a topic, find an idea, identify a literature gap, explore a re
image-to-svgReconstruct one PNG, JPG/JPEG, WebP, BMP, or TIFF image as a visually faithful, editable SVG with Calibri text and source-size render comparison. Use when the user asks YanShu or 研术台 to convert, vectorize, trace, or reproduce a raster image as SVG, especially for strict 1:1 visual replication.
paper-draftingDraft a complete, compilable LaTeX research paper with YanShu from completed experiments, code, tables, figures, method notes, and authentic references. Use when the user asks YanShu or 研术台 to write a first paper draft, turn finished experiments into a manuscript, create an arXiv draft, or prepare a
paper-reconstructionReconstruct an existing research paper with YanShu in one current-task execution from local TeX, BibTeX, PDF, and optional figure evidence. Use when the user asks YanShu or 研术台 to reconstruct, deeply refine, restructure, rewrite, or quality-check a complete paper while preserving its evidence and te
peer-reviewConduct an independent, evidence-grounded peer review of an academic manuscript with YanShu. Use when the user asks YanShu or 研术台 to review, referee, critique, stress-test, or assess a paper before submission, identify major and minor concerns, evaluate contribution, method, experiments, claims, pre
revision-auditVerify a revised manuscript, response letter or rebuttal, original manuscript, and diff against every reviewer comment with YanShu. Use when the user asks YanShu or 研术台 to audit a revision, check whether reviewer concerns were adequately addressed, validate claimed manuscript changes, assess a journ
revision-planningOrganize, merge, prioritize, and classify peer-review comments into an evidence-honest manuscript revision plan with YanShu. Use when the user asks YanShu or 研术台 to plan a revision, triage reviewer comments, decide which requests require experiments, identify minimum viable analyses, assess whether
scientific-figureDesign and generate one publication-ready scientific schematic with YanShu from a paper's TeX, optional PDF, and a reference image only when the user enables that option. Use when the user asks YanShu or 研术台 to draw a method overview, introduction figure, task definition, core mechanism detail, trai
yanshu-paper-workspaceruns: node ./scripts/node-launcher.cjs ./mcp/server.mjs --transport stdio
Talk like caveman. Cut filler. Keep technical details explicit.
22 skills · 1 command · 3 agents · hooks
Persistent memory for Claude Code: the next session starts where the last one left off.
29 skills · 1 MCP server · hooks
tasteskill 2.5.1: professional company websites at the figma.com and x.ai level, with one section per screen, real photographs, real product UI, effects kept for tech and gaming pages, and a real-browser checker and finish gate.
1 skill
Open-source AI job search that runs locally in your AI coding CLI: scan job portals, evaluate listings into a structured A-H report with a 1-5 score, tailor your CV, track applications
5 skills · hooks
Cross-session memory and token savings for coding agents.
6 skills · 1 agent · 1 MCP server · hooks
Package any website or local web build into a lightweight desktop app with Pake CLI.
1 skill
This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.