agentsclimarketplace

Run2 13f fund search

Skill cxcscmu/SkillLearnBench/skills/b2-self-feedback-gemini-3.1-pro-preview/financial-analysis/run2_13f-fund-search

How to correctly and robustly search for a fund manager's accession number in SEC 13F COVERPAGE.tsv.From its SKILL.md

Install
npx -y skills add cxcscmu/SkillLearnBench --skill run2_13f-fund-search

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

SKILL.md

1.3 KB, 299 tokens by cl100k_base, as published. Nobody here has run it

Overview

A critical first step in analyzing 13F data is identifying a fund manager's unique ACCESSION_NUMBER. Since names can be slightly misspelled or appended with corporate entities (like "LLC", "LP"), searching requires flexibility.

Usage Patterns

Fast Search with grep

The fastest and most reliable way to find the accession number is often via the command line using grep -i (case-insensitive search).

# Search for Renaissance Technologies in Q3 2025
grep -i "renaissance technologies" /root/2025-q3/COVERPAGE.tsv

Programmatic Search with pandas

When writing a script, use pandas with .str.contains rather than exact matching.

import pandas as pd

quarter = "2025-q3"
keywords = "BERKSHIRE HATHAWAY"

# Use dtype=str to avoid dropping leading zeros in IDs and to handle missing values cleanly
df = pd.read_csv(f"/root/{quarter}/COVERPAGE.tsv", sep="\t", dtype=str)

# Filter out amendments if you only want the original or final holdings report
df = df[df["REPORTTYPE"].str.contains("HOLDINGS REPORT", na=False)]

matches = df[df["FILINGMANAGER_NAME"].str.contains(keywords, case=False, na=False)]
print(matches[["ACCESSION_NUMBER", "FILINGMANAGER_NAME"]])

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.

Keep looking

Skills are one crate of 325,949. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.