maxqian888/cognia-next/latex-write

Claude Code and agent skill by maxqian888/cognia-next

Trust tier 2
  • community
LLM review pending
Source check pending

Checking that the SKILL.md file is still in the repository.

Draft or extend a LaTeX document from a writing brief through the Pi LaTeX Workbench, then build it to a draft PDF.

Install

npx vskill@latest install maxqian888/cognia-next/latex-write

SKILL.md

LaTeX: write

Follow the latex-workbench skill for tool mechanics (stateDir, project, ops files, apply → materialize → build, approvals). Then:

  1. latexwb_inspect the project. Reuse its existing structure, class,

preamble, macros and bibliography backend. Only when the user asks for a new document and no project exists, start from the approved template that matches the language and document type (for Chinese text a ctex-based template: zh-article, zh-thesis, zh-beamer) as the latex-workbench skill describes, and replace its placeholder text instead of appending.

  1. Read ${COGNIA_PLUGIN_ROOT}/vendor/resources/skills/latex-revise/references/guide.md

(“Draft or extend text”) and, before the first patch, ${COGNIA_PLUGIN_ROOT}/vendor/resources/skills/latex-project/references/guide.md.

  1. Write section by section: one governed proposal per section or file, so a

failing build points at a small region. Ask only for missing requirements that change the result; otherwise follow the document's conventions.

  1. Never invent references, data or results. Cite only entries that exist in

the project's bibliography or that the user supplied; mark illustrative numbers as illustrative. New equations, labels and citations are protected additions: in strict mode each such patch needs the operator's grant — batch them so one grant covers a coherent section, and give the user the exact approve command.

  1. After each applied proposal: materialize, latexwb_build, fix the first

causal error until the draft compiles, then latexwb_check_run with ruleset: "draft".

  1. Return the PDF (latexwb_artifact_save to where the user wants it) unless

the user asked for source only or for a formal release.

Apply this to the request the user made when invoking the skill.

Security

  • Security scanPassed, 96 of 100Pattern checks on the skill's files.
  • Model reviewNot runA model review is scheduled for the next rescan.

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

Repository stars
7
Repository forks
1
7-day trend score
0

Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/maxqian888/cognia-next/latex-write/badge)](https://verified-skill.com/skills/maxqian888/cognia-next/latex-write)

Something wrong with this skill or its page? Tell us.