cn-economics-thesis-workflow-skill
A nine-phase skill for a Chinese economics thesis: from topic and literature to regressions and docx layout.
- 94 ★ on the repository
- Last commit
Extended access
Why this level
- The agent runs Python or Stata scripts, reads your data and writes tables and documents into the project
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/cn-economics-thesis-workflow-skillDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- The agent runs Python or Stata scripts, reads your data and writes tables and documents into the project
This is third-party code. Review the repository files before installing.
What it does
The skill leads a Chinese-language empirical economics thesis through nine phases: environment and backend choice, topic selection, literature and theory framework, data and variables, baseline regression, endogeneity, robustness, mechanisms, heterogeneity and writing with layout. Each phase holds interaction nodes, a method list, acceptance criteria and reproducible script templates. Phases four to eight share one loop: method recommendation, node confirmation, variable readiness check, script writing and running, result reading and confirmation. Numbers in the text come only from script output and match files in `output/tables/`. The backend is a choice: Python with pyfixest, linearmodels and statsmodels (default) or Stata version 16 or later through mcp-for-stata. Output is tables in `.csv` and `.rtf`, charts and a thesis in `.docx`.
Who it is for. Master's and PhD economics students who write an empirical thesis in Chinese and want reproducible calculations.
Good fit when
- The topic is still open, while feasibility and literature remain unstudied.
- A baseline regression plus endogeneity and robustness checks are needed, with Python or Stata scripts.
- Numbers in the text must match the tables the scripts output.
- The thesis has to be formatted in `.docx` with three-line tables and citation substitution.
Not a fit when
- The thesis is in another language: the skill text and templates are written in Chinese.
- The study is not empirical: the phases assume regression analysis.
- You want a finished text with no author involvement: at each node the skill waits for confirmation.
Example request
Run phase 0 and choose the Python backend, then help me define a topic using enterprise panel data.Limitations
The skill materials are in Chinese and target empirical economics theses. The Python backend is the default, Stata has unique capabilities such as Goodman-Bacon decomposition, `rdrobust` and `csdid`, and needs Stata version 16 or newer and mcp-for-stata. The author specifies that for estimators missing in Python a simplified method is forbidden. Data and tables stay on your machine, but text and results enter the model context. State is stored in `.thesis_state.json`, and the result should be reviewed by a supervisor.
How to disable. Delete the skill folder from the agent skills directory or the rules file in OpenCode and Cursor. The created project, scripts and `.thesis_state.json` stay with you.
FAQ
How many phases are there?
Nine: from environment and topic to writing and layout. Phases four to eight follow one confirmation loop.
Which backend should I pick?
Python by default with pyfixest, linearmodels and statsmodels. Stata version 16 or later is needed for Goodman-Bacon, `rdrobust` and `csdid`.
Where do numbers in the text come from?
Only from script output, and values match the tables in `output/tables/`.
Which clients fit?
OpenCode, Claude Code, Codex, Cursor and others that understand skills or rule files. The OpenCode config is in `opencode.jsonc`.
Where is state stored?
In `.thesis_state.json`, which also records the chosen `analysis_backend`.
Discussion
No questions or reviews yet.
Related
Skills17 skills for economists: research ideas, literature review, Stata and R, LaTeX models, tables and Beamer slides
Claude Code skills for empirical economics: code and replication audit, bibliography checks, PDF reading and slide decks.
Auto-Empirical Research Skills
Auto-Empirical Research Skills (AERS)
A plugin marketplace for empirical social science research: causal inference, econometrics and manuscript prep for top journals
A skill for math modeling contests CUMCM, MCM/ICM and the Electrical Cup: 10 stages, three roles, five gates and a LaTeX paper
A skill that takes problem C of China's CUMCM mathematical modelling contest from data to a checked LaTeX paper.
A pack of 14 skills for public-data research in Chinese: a lead skill runs the project and 13 specialists cover market, competitors and valuation
README in short
The README lists work stages and outputs, the phase diagram, two backends and supporting appendices. Phases are described in separate files and code lives in appendices H and I.