Public tooling for the Factor Weave quant-factor data API — clients, add-ons, and integration recipes for the ~12,000 US-listed tickers covered by the platform.
Looking for a free key? Sign up at factorweave.com — 250 calls/day, no card.
| Path | What | Install / use |
|---|---|---|
python/ |
Official Python SDK — typed client with pandas / polars helpers, retry handling, tier-aware exceptions. Also installs an fw CLI. |
pip install factorweave · PyPI |
typescript/ |
Official TypeScript / JavaScript SDK — dual ESM+CJS, full types, native fetch, auto-retry on 429/5xx. Server-side Node 18+ / Bun / Deno / serverless. |
npm install @blazing-customs/factorweave · npm |
r/ |
Official R package — httr2-based, idiomatic data.frame returns, retry on 429/5xx. |
install.packages("factorweave", repos = "https://blazing-customs.r-universe.dev") · r-universe |
sheets/ |
Google Sheets add-on — exposes Factor Weave as spreadsheet custom functions (=FACTORWEAVE("AAPL","rsi")). |
Paste Code.gs into your Apps Script project. |
| Path | What |
|---|---|
generated/ |
openapi-generator clients from the live OpenAPI spec. Go, Rust, Ruby, PHP, Dart are committed; Java, C#, Kotlin, Swift are available on-demand via scripts/regenerate_clients.sh in the monorepo. |
| Path | What |
|---|---|
webhooks/ |
Alert-delivery templates — sample payload, Slack/Discord transformers (deployable as serverless functions), Zapier/Make/n8n setup guides + an importable n8n workflow, and a fire-test-payload.sh script for verifying your endpoint before going live. |
mcp-configs/ |
Drop-in MCP client configs for Claude Desktop, Cursor, Continue, Cline, Windsurf, plus a generic descriptor. Copy → replace fw_live_REPLACE_ME → restart your client. |
notebooks/ |
5 executable Jupyter notebooks covering the common workflows: first request · screening · similarity / peer set · leak-free backtest · regime conditioning. Run with or without an API key (demo fallback covers 8 sample tickers). |
postman/ |
Auto-generated Postman v2.1 collection (30+ requests, 21 groups) + environment file. Imports into Postman, Insomnia, Bruno, Thunder Client. |
More doorways will land here under additional top-level directories as we ship them.
# Python SDK
from factorweave import Client
client = Client(api_key="fw_live_…")
row = client.features("AAPL")[0]
top = client.top("mom", n=25).to_pandas()
hits = client.find_similar("AAPL", method="cosine", limit=10, min_lookback_days=30)
card = client.report_card("AAPL") # HOBBY+# fw CLI (ships with the Python package)
fw features AAPL
fw top mom -n 25
fw similar AAPL --method cosine// TypeScript SDK
import { FactorWeave } from '@blazing-customs/factorweave';
const fw = new FactorWeave({ apiKey: process.env.FACTORWEAVE_API_KEY });
const row = await fw.latestFeatures('AAPL');
const hits = await fw.similar('AAPL', { method: 'cosine', limit: 10 });
const card = await fw.reportCard('AAPL'); // HOBBY+# R package
library(factorweave)
client <- fw_client(api_key = "fw_live_…")
row <- fw_latest_features(client, "AAPL")
top <- fw_top(client, "mom", n = 25)
hits <- fw_similar(client, "AAPL", method = "cosine", limit = 10)# Google Sheets
=FACTORWEAVE("AAPL", "rsi,mom,comp_score") // spills across columns
=FW_TOP("mom", 25) // spills down a column
=FW_MARKET_CONTEXT() // current SPY-vol regime + dispersion + breadth
=FW_REPORT_CARD("AAPL") // per-ticker digest (HOBBY+)
- ~12,000 US-listed tickers
- Daily, point-in-time, leak-free
- ~28 factor columns per ticker-day (returns, momentum, mean-reversion, RSI, ATR%, realized vol, beta vs SPY, composite score, cross-sectional ranks)
- Forward-return labels (1d / 5d / 20d), leak-free
- SPY-vol regime tagging (low / mid / high)
- 32-dimensional regime-aware factor-state embeddings
- Top-K nearest analogues via cosine / DTW / label-aware / supervised PLS
- Daily factor dispersion, market breadth, regime transition odds
- Per-ticker risk-cluster tags (calm / normal / stressed)
Tier matrix is on the pricing page.
Factor Weave is a research substrate, not a return-prediction service. Our own leak-free probes show factor similarity does not forecast forward returns (cross-sectional information coefficient is statistically zero). Only risk-coherence — using factor analogues to forecast forward realized volatility — shows a meaningful signal (IC +0.062, t-stat +8.3 across 237 monthly observations 2005–2024). The full methodology and results are published at factorweave.com/research.html.
Use these tools the honest way: to screen, explore, and assemble research data. The thesis is yours.
- Main site · factorweave.com
- API docs · factorweave.com/api/docs
- OpenAPI spec · factorweave.com/api/openapi.json
- MCP setup · factorweave.com/mcp.html
- Integrations · factorweave.com/integrations.html
- Research note · factorweave.com/research.html
- llms.txt (LLM-readable index) · factorweave.com/llms.txt
MIT — see python/LICENSE.