autoscholar-cli 1.0.0 → 1.0.1
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/README.md +338 -141
- package/package.json +1 -1
package/README.md
CHANGED
|
@@ -1,21 +1,6 @@
|
|
|
1
|
-
<
|
|
2
|
-
<br/>
|
|
3
|
-
<strong>
|
|
4
|
-
<code> A u t o S c h o l a r C L I </code>
|
|
5
|
-
</strong>
|
|
6
|
-
<br/><br/>
|
|
7
|
-
<em>A quant researcher + economist + data scientist living inside your terminal.</em>
|
|
8
|
-
<br/><br/>
|
|
9
|
-
<a href="https://www.npmjs.com/package/autoscholar-cli"><img src="https://img.shields.io/npm/v/autoscholar-cli?color=0EA5E9&label=npm&style=flat-square" alt="npm version"></a>
|
|
10
|
-
<a href="https://www.npmjs.com/package/autoscholar-cli"><img src="https://img.shields.io/npm/dm/autoscholar-cli?color=06B6D4&style=flat-square" alt="downloads"></a>
|
|
11
|
-
<a href="https://github.com/autoscholar/autoscholar-cli/blob/main/LICENSE"><img src="https://img.shields.io/npm/l/autoscholar-cli?color=14B8A6&style=flat-square" alt="license"></a>
|
|
12
|
-
<img src="https://img.shields.io/badge/node-%3E%3D18-0EA5E9?style=flat-square" alt="node version">
|
|
13
|
-
<img src="https://img.shields.io/badge/python-%3E%3D3.9-0EA5E9?style=flat-square" alt="python version">
|
|
14
|
-
</p>
|
|
1
|
+
<div align="center">
|
|
15
2
|
|
|
16
|
-
|
|
17
|
-
|
|
18
|
-
**AutoScholar CLI** is an autonomous scientific research agent that generates **publication-ready academic papers** entirely from the command line. It orchestrates five AI agents to search literature, collect data, build datasets, run econometric analysis, and compile LaTeX PDFs — all without a browser or GUI.
|
|
3
|
+
<br/>
|
|
19
4
|
|
|
20
5
|
```
|
|
21
6
|
_ _ ____ _ _
|
|
@@ -23,190 +8,402 @@
|
|
|
23
8
|
/ _ \| | | | __/ _ \ \___ \ / __| '_ \ / _ \| |/ _` | '__|
|
|
24
9
|
/ ___ \ |_| | || (_) | ___) | (__| | | | (_) | | (_| | |
|
|
25
10
|
/_/ \_\__,_|\__\___/ |____/ \___|_| |_|\___/|_|\__,_|_|
|
|
11
|
+
```
|
|
12
|
+
|
|
13
|
+
### Autonomous Scientific Research Agent
|
|
14
|
+
|
|
15
|
+
**Generate publication-ready academic papers — entirely from your terminal.**
|
|
16
|
+
|
|
17
|
+
<br/>
|
|
18
|
+
|
|
19
|
+
[](https://www.npmjs.com/package/autoscholar-cli)
|
|
20
|
+
|
|
21
|
+
[](LICENSE)
|
|
22
|
+
|
|
23
|
+
[](https://nodejs.org)
|
|
24
|
+
|
|
25
|
+
[](https://python.org)
|
|
26
|
+
|
|
27
|
+
<br/>
|
|
26
28
|
|
|
27
|
-
|
|
29
|
+
[Install](#-install) • [Quick Start](#-quick-start) • [How It Works](#-how-it-works) • [Data Sources](#-data-sources) • [API Keys](#-api-keys)
|
|
30
|
+
|
|
31
|
+
<br/>
|
|
32
|
+
|
|
33
|
+
---
|
|
34
|
+
|
|
35
|
+
<br/>
|
|
36
|
+
|
|
37
|
+
<table>
|
|
38
|
+
<tr>
|
|
39
|
+
<td>
|
|
40
|
+
|
|
41
|
+
```
|
|
42
|
+
╭──────────────────────────────────────────────╮
|
|
43
|
+
│ Welcome to AutoScholar │
|
|
44
|
+
│ A quant researcher + economist + data │
|
|
45
|
+
│ scientist living inside your terminal. │
|
|
46
|
+
╰──────────────────────────────────────────────╯
|
|
47
|
+
|
|
48
|
+
◆ Pre-flight Checks
|
|
49
|
+
──────────────────────────────────────────────
|
|
50
|
+
|
|
51
|
+
✔ Python (python3) — 9 packages ready
|
|
52
|
+
✔ API keys — 5 configured
|
|
53
|
+
|
|
54
|
+
████████░░░░░░░░░░░░ 29% Gauss — Literature
|
|
55
|
+
████████████████░░░░ 71% Fisher — Analysis
|
|
56
|
+
████████████████████ 100% Finalizing
|
|
57
|
+
|
|
58
|
+
╔═══════════ Research Complete ════════════╗
|
|
59
|
+
║ ✔ Paper generated successfully ║
|
|
60
|
+
║ ║
|
|
61
|
+
║ Title Volatility Spillovers Between ║
|
|
62
|
+
║ Crypto and Commodities ║
|
|
63
|
+
║ Pages 72 ║
|
|
64
|
+
║ Figures 24 ║
|
|
65
|
+
║ Duration 18m 42s ║
|
|
66
|
+
╚════════════════════════════════════════════╝
|
|
28
67
|
```
|
|
29
68
|
|
|
30
|
-
|
|
69
|
+
</td>
|
|
70
|
+
</tr>
|
|
71
|
+
</table>
|
|
72
|
+
|
|
73
|
+
</div>
|
|
74
|
+
|
|
75
|
+
<br/>
|
|
76
|
+
|
|
77
|
+
## Why AutoScholar?
|
|
78
|
+
|
|
79
|
+
Most research tools help with *one step* — search papers, fetch data, or run a model. AutoScholar handles the **entire pipeline** autonomously:
|
|
80
|
+
|
|
81
|
+
| | Traditional Workflow | With AutoScholar |
|
|
82
|
+
|---|---|---|
|
|
83
|
+
| **Literature Review** | Hours of manual searching | 250M+ papers searched in minutes |
|
|
84
|
+
| **Data Collection** | Writing API scripts by hand | Auto-fetched from 12+ sources |
|
|
85
|
+
| **Analysis** | Manually coding models | GARCH, OLS, Panel — auto-generated |
|
|
86
|
+
| **Paper Writing** | Weeks of drafting | 8 sections written in LaTeX |
|
|
87
|
+
| **Figures** | One-by-one in matplotlib | 20+ publication-quality figures |
|
|
88
|
+
| **PDF** | Manual LaTeX compilation | Compiled automatically |
|
|
89
|
+
|
|
90
|
+
> **One command. One paper. Fully reproducible.**
|
|
91
|
+
|
|
92
|
+
<br/>
|
|
93
|
+
|
|
94
|
+
## 📦 Install
|
|
31
95
|
|
|
32
96
|
```bash
|
|
33
97
|
npm install -g autoscholar-cli
|
|
34
98
|
```
|
|
35
99
|
|
|
36
|
-
|
|
37
|
-
|
|
38
|
-
AutoScholar requires Python 3.9+ with the following packages:
|
|
100
|
+
Then install the Python scientific stack:
|
|
39
101
|
|
|
40
102
|
```bash
|
|
41
103
|
pip3 install pandas numpy scipy matplotlib seaborn scikit-learn statsmodels arch linearmodels
|
|
42
104
|
```
|
|
43
105
|
|
|
44
|
-
|
|
106
|
+
<details>
|
|
107
|
+
<summary><strong>Optional: LaTeX for PDF output</strong></summary>
|
|
108
|
+
<br/>
|
|
45
109
|
|
|
46
|
-
Install [TeX Live](https://www.tug.org/texlive/) or [MacTeX](https://www.tug.org/mactex/)
|
|
47
|
-
|
|
48
|
-
## Quick Start
|
|
110
|
+
Install [TeX Live](https://www.tug.org/texlive/) or [MacTeX](https://www.tug.org/mactex/) to compile papers to PDF. Without it, AutoScholar saves `.tex` source files that you can compile yourself.
|
|
49
111
|
|
|
50
112
|
```bash
|
|
51
|
-
#
|
|
52
|
-
|
|
113
|
+
# macOS
|
|
114
|
+
brew install --cask mactex-no-gui
|
|
53
115
|
|
|
54
|
-
#
|
|
55
|
-
|
|
116
|
+
# Ubuntu / Debian
|
|
117
|
+
sudo apt install texlive-full
|
|
56
118
|
```
|
|
57
119
|
|
|
58
|
-
|
|
59
|
-
|
|
60
|
-
## Commands
|
|
120
|
+
</details>
|
|
61
121
|
|
|
62
|
-
|
|
63
|
-
|---------|-------------|
|
|
64
|
-
| `autoscholar` | Interactive mode (default) |
|
|
65
|
-
| `autoscholar run` | Run the full research pipeline |
|
|
66
|
-
| `autoscholar config` | Manage API keys |
|
|
67
|
-
| `autoscholar outputs` | Browse generated papers |
|
|
68
|
-
| `autoscholar resume <id>` | Resume an incomplete project |
|
|
122
|
+
<br/>
|
|
69
123
|
|
|
70
|
-
|
|
124
|
+
## 🚀 Quick Start
|
|
71
125
|
|
|
72
126
|
```bash
|
|
127
|
+
# Interactive mode — guided menu
|
|
128
|
+
autoscholar
|
|
129
|
+
|
|
130
|
+
# Direct execution
|
|
131
|
+
autoscholar run --topic "Volatility spillovers between crypto and commodities 2015-2025"
|
|
132
|
+
|
|
133
|
+
# With all options
|
|
73
134
|
autoscholar run \
|
|
74
|
-
--topic "Factor momentum in US equities"
|
|
75
|
-
--assets "AAPL,MSFT,GOOG,AMZN"
|
|
76
|
-
--method "DCC-GARCH"
|
|
77
|
-
--hf
|
|
78
|
-
|
|
135
|
+
--topic "Factor momentum in US equities" \
|
|
136
|
+
--assets "AAPL,MSFT,GOOG,AMZN" \
|
|
137
|
+
--method "DCC-GARCH" \
|
|
138
|
+
--hf # high-frequency data
|
|
139
|
+
|
|
140
|
+
# Dry run — preview the plan without executing
|
|
141
|
+
autoscholar run --topic "..." --dry
|
|
79
142
|
```
|
|
80
143
|
|
|
81
|
-
|
|
144
|
+
> On first launch, a **setup wizard** walks you through API key configuration.
|
|
145
|
+
|
|
146
|
+
<br/>
|
|
147
|
+
|
|
148
|
+
## 🧠 How It Works
|
|
82
149
|
|
|
83
|
-
AutoScholar orchestrates **five specialized AI agents** through a 7-stage pipeline
|
|
150
|
+
AutoScholar orchestrates **five specialized AI agents** through a 7-stage pipeline, supervised by a rule-based **Governor Engine** that handles errors, retries, and quality control.
|
|
84
151
|
|
|
85
152
|
```
|
|
86
|
-
|
|
87
|
-
|
|
88
|
-
|
|
89
|
-
|
|
90
|
-
|
|
91
|
-
|
|
92
|
-
|
|
93
|
-
|
|
153
|
+
┌─────────────────────────────────────────────────────────────┐
|
|
154
|
+
│ │
|
|
155
|
+
│ 📋 Plan → 📚 Gauss → 📊 Turing → 🔧 Newton │
|
|
156
|
+
│ │
|
|
157
|
+
│ → 🔬 Fisher → ✍️ Write → 📄 Compile │
|
|
158
|
+
│ │
|
|
159
|
+
│ ┌───────────────────────────────────────────────────┐ │
|
|
160
|
+
│ │ Governor Engine — profiles, retries, confidence │ │
|
|
161
|
+
│ └───────────────────────────────────────────────────┘ │
|
|
162
|
+
└─────────────────────────────────────────────────────────────┘
|
|
94
163
|
```
|
|
95
164
|
|
|
165
|
+
<br/>
|
|
166
|
+
|
|
96
167
|
### The Five Agents
|
|
97
168
|
|
|
98
|
-
|
|
99
|
-
|
|
100
|
-
|
|
101
|
-
|
|
102
|
-
|
|
103
|
-
|
|
104
|
-
|
|
169
|
+
<table>
|
|
170
|
+
<tr>
|
|
171
|
+
<td width="100" align="center"><h3>Euler</h3><sub>Orchestrator</sub></td>
|
|
172
|
+
<td>Plans the research strategy, coordinates all agents, decides methodology, writes all 8 paper sections, and compiles the final PDF. The brain of the system.</td>
|
|
173
|
+
</tr>
|
|
174
|
+
<tr>
|
|
175
|
+
<td width="100" align="center"><h3>Gauss</h3><sub>Literature</sub></td>
|
|
176
|
+
<td>Searches <strong>250M+ academic papers</strong> via OpenAlex, arXiv, and CrossRef. Expands citation networks. Identifies research gaps using 6 detection engines (method, dataset, temporal, contradiction, limitation, cross-disciplinary).</td>
|
|
177
|
+
</tr>
|
|
178
|
+
<tr>
|
|
179
|
+
<td width="100" align="center"><h3>Turing</h3><sub>Data</sub></td>
|
|
180
|
+
<td>Collects raw data from <strong>12+ APIs</strong> — FMP (stocks, fundamentals, SEC filings), EODHD (historical, intraday, options), FRED (840K+ macro series), World Bank, IMF. Saves everything with metadata for reproducibility.</td>
|
|
181
|
+
</tr>
|
|
182
|
+
<tr>
|
|
183
|
+
<td width="100" align="center"><h3>Newton</h3><sub>Dataset</sub></td>
|
|
184
|
+
<td>Cleans missing values, merges multiple data sources, constructs derived variables (returns, lags, rolling stats), handles outliers (winsorization), and builds the final analysis-ready dataset.</td>
|
|
185
|
+
</tr>
|
|
186
|
+
<tr>
|
|
187
|
+
<td width="100" align="center"><h3>Fisher</h3><sub>Analysis</sub></td>
|
|
188
|
+
<td>Estimates econometric models (GARCH, OLS, Panel FE, IV/GMM), generates <strong>20+ publication-quality figures</strong>, creates LaTeX tables with significance stars, and runs diagnostic tests (VIF, Durbin-Watson, heteroskedasticity).</td>
|
|
189
|
+
</tr>
|
|
190
|
+
</table>
|
|
191
|
+
|
|
192
|
+
<br/>
|
|
105
193
|
|
|
106
194
|
### Governor Engine
|
|
107
195
|
|
|
108
|
-
A
|
|
109
|
-
|
|
110
|
-
- Automatic **profile switching** (academic figures, debug recovery, LaTeX fallback)
|
|
111
|
-
- **Retry budgets** (5 Python, 3 data, 3 LaTeX retries)
|
|
112
|
-
- **Confidence scoring** that adapts to errors and successes
|
|
196
|
+
A zero-LLM decision engine monitors every step:
|
|
113
197
|
|
|
114
|
-
|
|
198
|
+
| Feature | Description |
|
|
199
|
+
|---------|-------------|
|
|
200
|
+
| **Profile switching** | Auto-switches to `academic_figures` during Fisher, `jf_rewrite` during writing, `debug_recovery` on Python errors, `latex_fallback` on compilation errors |
|
|
201
|
+
| **Retry budgets** | 5 Python retries, 3 data retries, 3 LaTeX retries — then graceful finalization |
|
|
202
|
+
| **Confidence scoring** | Starts at 0.8, adjusts on every success (+0.02) and failure (-0.05 to -0.25) |
|
|
203
|
+
| **Error pattern matching** | Detects `ModuleNotFoundError`, `SingularMatrix`, `MemoryError`, etc. and injects targeted recovery directives |
|
|
115
204
|
|
|
116
|
-
|
|
117
|
-
|--------|----------|-------------|
|
|
118
|
-
| **OpenAlex** | 250M+ academic papers | No |
|
|
119
|
-
| **arXiv** | Preprints (finance, econ, stat) | No |
|
|
120
|
-
| **CrossRef** | 140M+ articles, DOI resolution | No |
|
|
121
|
-
| **FMP** | 230+ financial data endpoints | Yes |
|
|
122
|
-
| **EODHD** | Historical prices, intraday, options | Yes |
|
|
123
|
-
| **FRED** | 840,000+ macroeconomic series | Yes |
|
|
124
|
-
| **World Bank** | 16,000+ development indicators | No |
|
|
125
|
-
| **IMF** | Global Financial Stability data | No |
|
|
126
|
-
| **Firecrawl** | Web scraping & extraction | Yes |
|
|
127
|
-
| **Serper / SerpAPI** | Web & Google Scholar search | Yes |
|
|
128
|
-
| **Tavily** | AI-powered search | Yes |
|
|
129
|
-
| **Exa** | Semantic neural search | Yes |
|
|
205
|
+
<br/>
|
|
130
206
|
|
|
131
|
-
##
|
|
207
|
+
## 🖥️ Commands
|
|
132
208
|
|
|
133
|
-
|
|
209
|
+
| Command | Description |
|
|
210
|
+
|:--------|:------------|
|
|
211
|
+
| `autoscholar` | Launch interactive mode with guided menu |
|
|
212
|
+
| `autoscholar run` | Execute the full research pipeline |
|
|
213
|
+
| `autoscholar config` | Add, edit, or validate API keys |
|
|
214
|
+
| `autoscholar outputs` | Browse all generated papers and figures |
|
|
215
|
+
| `autoscholar resume <id>` | Resume an interrupted project |
|
|
216
|
+
|
|
217
|
+
### Run Flags
|
|
218
|
+
|
|
219
|
+
| Flag | Description | Example |
|
|
220
|
+
|:-----|:------------|:--------|
|
|
221
|
+
| `--topic` | Research topic (required) | `"Crypto volatility spillovers"` |
|
|
222
|
+
| `--assets` | Comma-separated tickers | `"BTC,ETH,WTI,GOLD"` |
|
|
223
|
+
| `--method` | Econometric method | `"DCC-GARCH"`, `"OLS"`, `"VAR"` |
|
|
224
|
+
| `--hf` | Use high-frequency intraday data | — |
|
|
225
|
+
| `--dry` | Show execution plan only | — |
|
|
226
|
+
|
|
227
|
+
<br/>
|
|
228
|
+
|
|
229
|
+
## 🌐 Data Sources
|
|
230
|
+
|
|
231
|
+
<table>
|
|
232
|
+
<tr><th align="left">Source</th><th align="left">Coverage</th><th align="center">Free</th></tr>
|
|
233
|
+
<tr><td><strong>OpenAlex</strong></td><td>250M+ academic papers, citation networks</td><td align="center">✅</td></tr>
|
|
234
|
+
<tr><td><strong>arXiv</strong></td><td>Preprints — finance, econ, stat, CS</td><td align="center">✅</td></tr>
|
|
235
|
+
<tr><td><strong>CrossRef</strong></td><td>140M+ articles, DOI resolution</td><td align="center">✅</td></tr>
|
|
236
|
+
<tr><td><strong>World Bank</strong></td><td>16,000+ development indicators, 200+ countries</td><td align="center">✅</td></tr>
|
|
237
|
+
<tr><td><strong>IMF</strong></td><td>WEO, Global Financial Stability, BOP</td><td align="center">✅</td></tr>
|
|
238
|
+
<tr><td colspan="3"></td></tr>
|
|
239
|
+
<tr><td><strong>FMP</strong></td><td>230+ endpoints — fundamentals, market data, SEC, ESG</td><td align="center">🔑</td></tr>
|
|
240
|
+
<tr><td><strong>EODHD</strong></td><td>Historical prices, intraday, options with Greeks</td><td align="center">🔑</td></tr>
|
|
241
|
+
<tr><td><strong>FRED</strong></td><td>840,000+ macroeconomic time series</td><td align="center">🔑</td></tr>
|
|
242
|
+
<tr><td><strong>Firecrawl</strong></td><td>Web scraping, PDF extraction, structured data</td><td align="center">🔑</td></tr>
|
|
243
|
+
<tr><td><strong>Serper</strong></td><td>Google Search API</td><td align="center">🔑</td></tr>
|
|
244
|
+
<tr><td><strong>SerpAPI</strong></td><td>Google Scholar search</td><td align="center">🔑</td></tr>
|
|
245
|
+
<tr><td><strong>Tavily</strong></td><td>AI-powered search with generated answers</td><td align="center">🔑</td></tr>
|
|
246
|
+
<tr><td><strong>Exa</strong></td><td>Semantic neural search</td><td align="center">🔑</td></tr>
|
|
247
|
+
</table>
|
|
248
|
+
|
|
249
|
+
<sub>✅ = No API key needed 🔑 = API key required</sub>
|
|
250
|
+
|
|
251
|
+
<br/>
|
|
252
|
+
|
|
253
|
+
## 🔐 API Keys
|
|
254
|
+
|
|
255
|
+
Keys are stored locally in `~/.autoscholar/.env`. Manage them anytime:
|
|
134
256
|
|
|
135
257
|
```bash
|
|
136
258
|
autoscholar config
|
|
137
259
|
```
|
|
138
260
|
|
|
139
|
-
| Key |
|
|
140
|
-
|
|
141
|
-
| `ANTHROPIC_API_KEY` | **
|
|
142
|
-
| `FMP_API_KEY` | Recommended |
|
|
143
|
-
| `EODHD_API_KEY` | Recommended |
|
|
144
|
-
| `FRED_API_KEY` | Recommended |
|
|
145
|
-
| `SERPER_API_KEY` | Optional | Web search |
|
|
146
|
-
| `SERPAPI_API_KEY` | Optional | Google Scholar |
|
|
147
|
-
| `FIRECRAWL_API_KEY` | Optional |
|
|
148
|
-
| `TAVILY_API_KEY` | Optional | AI search |
|
|
149
|
-
| `EXA_API_KEY` | Optional | Semantic search |
|
|
261
|
+
| Key | Status | What it unlocks |
|
|
262
|
+
|:----|:------:|:----------------|
|
|
263
|
+
| `ANTHROPIC_API_KEY` | **Required** | Powers all 5 AI agents (Claude) |
|
|
264
|
+
| `FMP_API_KEY` | Recommended | Stocks, fundamentals, ratios, SEC filings, ESG |
|
|
265
|
+
| `EODHD_API_KEY` | Recommended | Deep historical prices, intraday, options Greeks |
|
|
266
|
+
| `FRED_API_KEY` | Recommended | GDP, CPI, unemployment, rates — 840K+ series |
|
|
267
|
+
| `SERPER_API_KEY` | Optional | Web search for broader research context |
|
|
268
|
+
| `SERPAPI_API_KEY` | Optional | Google Scholar for academic paper discovery |
|
|
269
|
+
| `FIRECRAWL_API_KEY` | Optional | Scrape web pages, extract datasets from PDFs |
|
|
270
|
+
| `TAVILY_API_KEY` | Optional | AI-powered search with summarization |
|
|
271
|
+
| `EXA_API_KEY` | Optional | Semantic similarity search for papers & data |
|
|
150
272
|
|
|
151
|
-
|
|
273
|
+
<br/>
|
|
152
274
|
|
|
153
|
-
|
|
275
|
+
## 📁 Output Structure
|
|
276
|
+
|
|
277
|
+
Every project is fully self-contained and reproducible:
|
|
154
278
|
|
|
155
279
|
```
|
|
156
|
-
~/.autoscholar/projects/
|
|
157
|
-
|
|
158
|
-
|
|
159
|
-
|
|
160
|
-
|
|
161
|
-
|
|
162
|
-
|
|
163
|
-
|
|
164
|
-
|
|
165
|
-
|
|
166
|
-
|
|
167
|
-
|
|
168
|
-
|
|
169
|
-
|
|
170
|
-
|
|
171
|
-
|
|
172
|
-
|
|
173
|
-
|
|
174
|
-
|
|
175
|
-
|
|
280
|
+
~/.autoscholar/projects/AS-20250410-x8k2m1/
|
|
281
|
+
│
|
|
282
|
+
├── data/ ← Raw & processed datasets
|
|
283
|
+
│ ├── fmp_prices.csv
|
|
284
|
+
│ ├── fred_macro.csv
|
|
285
|
+
│ └── final_dataset.csv
|
|
286
|
+
│
|
|
287
|
+
├── code/ ← All generated Python scripts
|
|
288
|
+
│ ├── newton_build_dataset.py
|
|
289
|
+
│ └── fisher_analysis.py
|
|
290
|
+
│
|
|
291
|
+
├── output/ ← Publication artifacts
|
|
292
|
+
│ ├── paper.tex
|
|
293
|
+
│ ├── paper.pdf
|
|
294
|
+
│ ├── references.bib
|
|
295
|
+
│ └── figures/
|
|
296
|
+
│ ├── correlation_matrix.png
|
|
297
|
+
│ ├── time_series.png
|
|
298
|
+
│ ├── garch_volatility.png
|
|
299
|
+
│ ├── ols_diagnostics.png
|
|
300
|
+
│ ├── scatter_matrix.png
|
|
301
|
+
│ └── ... (20+ figures)
|
|
302
|
+
│
|
|
303
|
+
├── logs/ ← Full execution logs
|
|
304
|
+
│ └── run_2025-04-10T14-30-00.log
|
|
305
|
+
│
|
|
306
|
+
└── meta.json ← Project metadata & state
|
|
176
307
|
```
|
|
177
308
|
|
|
178
|
-
|
|
309
|
+
<br/>
|
|
310
|
+
|
|
311
|
+
## 🔬 Supported Methods
|
|
312
|
+
|
|
313
|
+
<details>
|
|
314
|
+
<summary><strong>Econometrics</strong></summary>
|
|
315
|
+
<br/>
|
|
179
316
|
|
|
180
|
-
|
|
181
|
-
|
|
317
|
+
| Method | Description |
|
|
318
|
+
|--------|-------------|
|
|
319
|
+
| OLS / Robust SE | Ordinary least squares with HC1-HC3 standard errors |
|
|
320
|
+
| IV / 2SLS | Instrumental variables estimation |
|
|
321
|
+
| GMM | Generalized method of moments |
|
|
322
|
+
| Panel FE / RE | Fixed and random effects with Hausman test |
|
|
323
|
+
| Fama-MacBeth | Two-pass cross-sectional regression |
|
|
324
|
+
| GARCH / DCC-GARCH | Volatility modeling and dynamic correlations |
|
|
325
|
+
| VAR / VECM | Vector autoregression and error correction |
|
|
326
|
+
| Granger Causality | Temporal precedence testing |
|
|
327
|
+
| Local Projections | Impulse response estimation |
|
|
182
328
|
|
|
183
|
-
|
|
184
|
-
t-tests, ANOVA, Chi-square, Mann-Whitney, Kruskal-Wallis, Bootstrap Confidence Intervals
|
|
329
|
+
</details>
|
|
185
330
|
|
|
186
|
-
|
|
187
|
-
|
|
331
|
+
<details>
|
|
332
|
+
<summary><strong>Statistics</strong></summary>
|
|
333
|
+
<br/>
|
|
188
334
|
|
|
189
|
-
|
|
335
|
+
| Method | Description |
|
|
336
|
+
|--------|-------------|
|
|
337
|
+
| t-tests | One-sample, two-sample, paired |
|
|
338
|
+
| ANOVA | One-way and two-way analysis of variance |
|
|
339
|
+
| Chi-square | Independence and goodness-of-fit |
|
|
340
|
+
| Mann-Whitney | Non-parametric rank test |
|
|
341
|
+
| Kruskal-Wallis | Non-parametric ANOVA |
|
|
342
|
+
| Bootstrap CI | Confidence intervals via resampling |
|
|
190
343
|
|
|
191
|
-
|
|
192
|
-
- **Python** >= 3.9
|
|
193
|
-
- **Anthropic API key** (required)
|
|
194
|
-
- **LaTeX** (optional, for PDF output)
|
|
344
|
+
</details>
|
|
195
345
|
|
|
196
|
-
|
|
346
|
+
<details>
|
|
347
|
+
<summary><strong>Machine Learning</strong></summary>
|
|
348
|
+
<br/>
|
|
197
349
|
|
|
198
|
-
|
|
350
|
+
| Method | Description |
|
|
351
|
+
|--------|-------------|
|
|
352
|
+
| Random Forest | Ensemble tree-based classification/regression |
|
|
353
|
+
| XGBoost | Gradient boosting with regularization |
|
|
354
|
+
| LightGBM | Fast gradient boosting for large datasets |
|
|
355
|
+
| SVM / SVR | Support vector machines |
|
|
356
|
+
| ElasticNet | L1 + L2 regularized regression |
|
|
357
|
+
| PCA | Dimensionality reduction |
|
|
358
|
+
| k-means | Unsupervised clustering |
|
|
359
|
+
| SHAP | Model interpretability and feature importance |
|
|
199
360
|
|
|
200
|
-
>
|
|
361
|
+
</details>
|
|
362
|
+
|
|
363
|
+
<br/>
|
|
364
|
+
|
|
365
|
+
## ⚙️ Requirements
|
|
366
|
+
|
|
367
|
+
| Dependency | Version | Required |
|
|
368
|
+
|:-----------|:--------|:--------:|
|
|
369
|
+
| Node.js | >= 18 | Yes |
|
|
370
|
+
| Python | >= 3.9 | Yes |
|
|
371
|
+
| Anthropic API key | — | Yes |
|
|
372
|
+
| LaTeX (TeX Live / MacTeX) | 2024+ | Optional |
|
|
373
|
+
|
|
374
|
+
<br/>
|
|
375
|
+
|
|
376
|
+
## 💡 Philosophy
|
|
377
|
+
|
|
378
|
+
<br/>
|
|
379
|
+
|
|
380
|
+
<div align="center">
|
|
381
|
+
|
|
382
|
+
> *A quant researcher + economist + data scientist*
|
|
383
|
+
> *living inside your terminal.*
|
|
384
|
+
|
|
385
|
+
<br/>
|
|
386
|
+
|
|
387
|
+
| | |
|
|
388
|
+
|:---:|:---|
|
|
389
|
+
| **Autonomous** | End-to-end paper generation — no manual steps |
|
|
390
|
+
| **Reproducible** | All code, data, and logs saved per project |
|
|
391
|
+
| **Modular** | Five specialized agents with clean interfaces |
|
|
392
|
+
| **Local-first** | Runs on your machine — no cloud dependency |
|
|
393
|
+
| **Premium** | Beautiful terminal UX inspired by Vercel & Stripe |
|
|
394
|
+
|
|
395
|
+
</div>
|
|
396
|
+
|
|
397
|
+
<br/>
|
|
398
|
+
|
|
399
|
+
---
|
|
201
400
|
|
|
202
|
-
|
|
401
|
+
<div align="center">
|
|
402
|
+
<br/>
|
|
203
403
|
|
|
204
|
-
|
|
205
|
-
- **Reproducible** — all code, data, and logs saved per project
|
|
206
|
-
- **Modular** — five specialized agents with clean interfaces
|
|
207
|
-
- **Local-first** — everything runs on your machine, no cloud dependency
|
|
208
|
-
- **Premium** — beautiful terminal UX inspired by Vercel and Stripe CLIs
|
|
404
|
+
**Built with** [Claude](https://anthropic.com) · [OpenAlex](https://openalex.org) · [arXiv](https://arxiv.org) · [FRED](https://fred.stlouisfed.org)
|
|
209
405
|
|
|
210
|
-
|
|
406
|
+
<sub>MIT License</sub>
|
|
211
407
|
|
|
212
|
-
|
|
408
|
+
<br/>
|
|
409
|
+
</div>
|
package/package.json
CHANGED