Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/cxcscmu/skilllearnbench/holdings-comparisonnpx skills add cxcscmu/SkillLearnBench --skill holdings-comparisongit clone --depth 1 https://github.com/cxcscmu/SkillLearnBenchWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/cxcscmu/skilllearnbench/holdings-comparison)<a href="https://agentmods.dev/skills/cxcscmu/skilllearnbench/holdings-comparison"><img src="https://agentmods.dev/badge/skills/cxcscmu/skilllearnbench/holdings-comparison.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00031 | $0.00850 |
| Opus 5 | $0.00015 | $0.00425 |
| Sonnet 5 | $0.00006 | $0.00170 |
| Haiku 4.5 | $0.00003 | $0.00085 |
Grade A, and why
holdings-comparison scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 69 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Holdings Comparison Skill
Overview
To compare holdings across quarters, load INFOTABLE for both quarters, filter by accession number, then compute the difference in VALUE or SSHPRNAMT.
Step-by-step
1. Get accession numbers for both quarters
import pandas as pd
q2_cover = pd.read_csv("/root/2025-q2/COVERPAGE.tsv", sep="\t", dtype=str)
q3_cover = pd.read_csv("/root/2025-q3/COVERPAGE.tsv", sep="\t", dtype=str)
q2_acc = q2_cover[q2_cover["FILINGMANAGER_NAME"].str.contains("berkshire", case=False, na=False)]["ACCESSION_NUMBER"].iloc[0]
q3_acc = q3_cover[q3_cover["FILINGMANAGER_NAME"].str.contains("berkshire", case=False, na=False)]["ACCESSION_NUMBER"].iloc[0]
2. Load holdings for both quarters
q2_info = pd.read_csv("/root/2025-q2/INFOTABLE.tsv", sep="\t", dtype=str)
q3_info = pd.read_csv("/root/2025-q3/INFOTABLE.tsv", sep="\t", dtype=str)
q2_holdings = q2_info[q2_info["ACCESSION_NUMBER"] == q2_acc].copy()
q3_holdings = q3_info[q3_info["ACCESSION_NUMBER"] == q3_acc].copy()
q2_holdings["VALUE"] = pd.to_numeric(q2_holdings["VALUE"], errors="coerce").fillna(0)
q3_holdings["VALUE"] = pd.to_numeric(q3_holdings["VALUE"], errors="coerce").fillna(0)
3. Merge and compute changes
# Aggregate by CUSIP (a fund may have multiple entries per stock for different share types)
q2_agg = q2_holdings.groupby("CUSIP")["VALUE"].sum().reset_index().rename(columns={"VALUE": "VALUE_Q2"})
q3_agg = q3_holdings.groupby("CUSIP")["VALUE"].sum().reset_index().rename(columns={"VALUE": "VALUE_Q3"})
merged = pd.merge(q2_agg, q3_agg, on="CUSIP", how="outer").fillna(0)
merged["CHANGE"] = merged["VALUE_Q3"] - merged["VALUE_Q2"]
# Top 5 increased positions
top5 = merged.sort_values("CHANGE", ascending=False).head(5)
print(top5[["CUSIP", "VALUE_Q2", "VALUE_Q3", "CHANGE"]])
4. Find top investors in a specific stock
# Find all funds holding a specific CUSIP in Q3
palantir_cusip = "69608A108" # example
holders = q3_info[q3_info["CUSIP"] == palantir_cusip].copy()
holders["VALUE"] = pd.to_numeric(holders["VALUE"], errors="coerce").fillna(0)
# Aggregate by accession number and merge with fund names
holders_agg = holders.groupby("ACCESSION_NUMBER")["VALUE"].sum().reset_index()
holders_with_names = holders_agg.merge(q3_cover[["ACCESSION_NUMBER", "FILINGMANAGER_NAME"]], on="ACCESSION_NUMBER")
top3 = holders_with_names.sort_values("VALUE", ascending=False).head(3)
print(top3[["FILINGMANAGER_NAME", "VALUE"]])
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 69 lines · 31 tokens per session scan A 2af20bd40bca
holdings-comparison is a skill published in the GitHub repository cxcscmu/SkillLearnBench (83 stars, last pushed 1mo ago), licensed MIT. It adds 31 tokens to every session and 850 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
portfolio
Cross-chain DeFi portfolio discovery, rebalancing suggestions, and NEAR Intent construction. Activates when the user pastes a wallet address or asks about yield/positions/rebalancing. Bootstraps a per-user "portfolio" project, aggregates positions across all the user's addresses inside one project, and offers a…
trader-setup
One-time onboarding for the financial trader workflow — real-time alerts, position-aware relevance, decision journaling with outcome tracking. After successful setup this skill is excluded from selection until the marker file is deleted.
paired-reversal-cancellation
Reference for paired-reversal handling on GL batch tapes: an RV row whose ref-trace points at an earlier row cancels BOTH legs (the RV and the row it references) — but ONLY when the two value-dates fall within the shop's reversal settlement window; an RV that references a too-old posting is NOT a cancellation and…
triangulated-fx-rates
Reference for multi-hop / triangulated currency conversion on batch rate tapes. When a rate row carries an explicit "via" currency, the row's rate is only one leg of the conversion, and the via currency may ITSELF be quoted through another via — so the effective rate is the product of every leg, resolved by walking…
timeseries-detrending
Tools and techniques for detrending time series data in macroeconomic analysis. Use when working with economic time series that need to be decomposed into trend and cyclical components. Covers HP filter, log transformations for growth series, and correlation analysis of business cycles.
fuzzy-name-search
This skill includes search capability in 13F, such as fuzzy search a fund information using possibly inaccurate name, or fuzzy search a stock cusip info using its name.