FinAI Research Workflow skill
Skills and a pipeline for economics and finance research: from a topic to a verifiable LaTeX paper draft.
- 109 ★ on the repository
- Last commit
- MIT
Extended access
Why this level
- The pipeline downloads data from external sources and spends model tokens.
- Some data directories are flagged as legally risky.
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/finai-research-workflowDetects 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 pipeline downloads data from external sources and spends model tokens.
- Some data directories are flagged as legally risky.
This is third-party code. Review the repository files before installing.
What it does
FinAI Research Workflow takes a study from a one-line topic to a LaTeX draft. Eighteen skills cover idea discovery, literature review, novelty check, experiment design, data acquisition, writing, figures, LaTeX compilation and review loops. Under the hood are 58 econometric modules: DID, event study, synthetic control, instrumental variables, RDD and more. The pipeline starts with `finai-pipeline --topic "..." --use-hitl`, and every stage needs your approval. Data comes from 43 MCP directories (financial statements, macroeconomics, academic papers). Synthetic data is not used without your consent. Templates cover more than 30 journals.
Who it is for. Economists and finance PhD students who need a reproducible pipeline with econometrics and journal templates rather than a one-off draft.
Good fit when
- A topic is stated in one sentence and a literature review with a novelty check is needed.
- A paper needs DID estimates with parallel-trend tests and robustness checks.
- A draft must be formatted to a specific journal template.
- Every stage should pass manual approval.
Not a fit when
- You want trading signals or investment advice: this is a research tool.
- You have no model API key: by default the pipeline exits with an error without `DEEPSEEK_API_KEY`.
- Your field is outside economics and finance.
Example request
Run a study on emissions trading and green innovation: review, novelty check and a DID experiment plan.Limitations
A model key is required (DeepSeek in the example, Anthropic and OpenAI supported) along with Python 3.10 to 3.13. Three data directories are flagged as legally risky and connect only with explicit consent; Chinese literature is limited. Methods with external dependencies such as `linearmodels` fail with a clear error when the backend is missing. The result remains a draft: authorship, data and conclusions are checked by the researcher. It is not a trading tool and gives no investment recommendations.
How to disable. Delete the virtual environment with the `finai-research-workflow` package and the copied skills from the agent's skills directory.
FAQ
What does `--use-hitl` mean?
It enables human checkpoints: each stage waits for your approval, so the paper is not written autonomously.
Which keys are needed?
A model key. In the example it is `DEEPSEEK_API_KEY`; `ANTHROPIC_API_KEY` and `OPENAI_API_KEY` are also listed. Some data directories need their own keys.
Which methods are included?
58 modules: DID, event study, Bacon decomposition, synthetic control, IV, panel GMM, RDD, mediation and more.
Does it work with Claude Code and Codex?
Skills live in `.agents/skills/` and `.claude/skills/`. Codex has `AGENTS.md`, and Claude Code and Cursor use the same skills.
Where does the data come from?
From 43 MCP directories: Chinese and US company statements, macro data (FRED, World Bank, IMF) and OpenAlex. Three legally risky directories connect only with consent.
Discussion
No questions or reviews yet.
Related
Skills17 skills for economists: research ideas, literature review, Stata and R, LaTeX models, tables and Beamer slides
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
26 skills and four reviewer agents for science in Claude Code: figure and statistics review, literature search, work handoff.
A skill plugin for Calkit projects that builds a reproducible research pipeline and traces where manuscript numbers came from.
Claude Code skills for empirical economics: code and replication audit, bibliography checks, PDF reading and slide decks.
Step-by-step causal inference: 14 skills pick a method, check assumptions, write R or Python code and stress-test the result.
README in short
Explains the pipeline principles (human in the loop, data verification), quick start via PyPI, the skill list, data directories and supported journal templates.