13f-analyzer
Analyze SEC 13F holdings (INFOTABLE) to compute a fund’s AUM, stock-holding count, quarter-over-quarter changes, and top holders of a given stock.
- License 许可
- Apache-2.0
- 最近更新
- 2026-11-14
- Stars
- ★ 0
- Forks
- 1
01 前言 Front Matter
- name
- 13f-analyzer
- description
- Analyze SEC 13F holdings (INFOTABLE) to compute a fund’s AUM, stock-holding count, quarter-over-quarter changes, and top holders of a given stock.
02 资源文件 2
- 脚本scripts/holding_analysis.py1.4 KB
- 脚本scripts/one_fund_analysis.py4.2 KB
03 SKILL.md 正文
Steps
- From the workspace root (the directory that contains
13f-analyzer/), verify inputs and helper scripts exist so you don’t waste time “finding” them:- Data:
/root/2025-q2/INFOTABLE.tsv,/root/2025-q2/COVERPAGE.tsv,/root/2025-q3/INFOTABLE.tsv,/root/2025-q3/COVERPAGE.tsv - Scripts:
ls 13f-analyzer/scripts/
- Data:
- To get one fund’s summary in a single quarter (AUM and stock count), run:
Interpret the printed fields precisely:python3 13f-analyzer/scripts/one_fund_analysis.py \ --accession_number <ACCESSION_NUMBER> \ --quarter 2025-q3Total AUM= sum ofVALUEacross all rows for that accession number (use this for “AUM”).Number of stock holdings= row-count after the helper’sTITLEOFCLASSstock filter. When asked “how many stocks”, use this field (notTotal number of holdings, not unique CUSIPs, not COVERPAGE totals).
- To compare a fund between quarters and get the biggest dollar-value increases (top buys), run:
For a request “ranked by dollar value increase”, take the required number of CUSIPs from the beginning of the helper’spython3 13f-analyzer/scripts/one_fund_analysis.py \ --quarter 2025-q3 \ --accession_number <Q3_ACCESSION_NUMBER> \ --baseline_quarter 2025-q2 \ --baseline_accession_number <Q2_ACCESSION_NUMBER>Top 10 Buyssection, preserving order (don’t re-rank by shares or % change). Verify you return exactly k unique CUSIP strings. - To rank which funds hold a given stock CUSIP by notional value, first follow the helper’s file-path convention (it reads
/root/INFOTABLE.tsv, not/root/<quarter>/INFOTABLE.tsv):
The output is ranked by summedln -sf /root/2025-q3/INFOTABLE.tsv /root/INFOTABLE.tsv python3 13f-analyzer/scripts/holding_analysis.py \ --cusip <CUSIP> \ --quarter 2025-q3 \ --topk 10VALUEperACCESSION_NUMBER. Convert the top accession numbers to manager names usingfuzzy-name-search’s exact accession-number lookup. - When the task requires a deliverable file, write the final numbers/lists into
/root/answers.json(absolute path) using the required schema, and verify it exists (e.g.,ls -l /root/answers.json).
Expected Result
You can (a) extract a quarter’s fund AUM and stock-holding count from the helper’s printed summary, (b) extract top buy CUSIPs from the helper’s Q2→Q3 comparison output, (c) rank top holders of a CUSIP in Q3 and map accession numbers to manager names, and (d) populate /root/answers.json with the required fields.
04 同仓库更多 Skills
video-frame-extraction
Extract frames from video files and save them as images using OpenCV
openai-vision
Analyze images and multi-frame sequences using OpenAI GPT vision models
image-ocr
Extract text content from images using Tesseract OCR via Python, including structured field extraction from receipts and forms
video-frame-extraction
Extract frames from video files and save them as images using OpenCV
openai-vision
Analyze images and multi-frame sequences using OpenAI GPT vision models
image-ocr
Extract text content from images using Tesseract OCR via Python, including structured field extraction from receipts and forms