cxcscmu/SkillLearnBench - 第 20 页

python-docx-nested-table-recursion

Detailed guidance on recursively processing nested tables in python-docx documents. Use this when a Word template contains tables inside table cells and you need to reach all paragraphs.

原文语言：英语

pdf-text-extraction-for-classification

How to extract text from PDF files for content-based classification. Use this when you need to read PDF content to determine what subject/topic a paper belongs to. Covers both text-based and scanned PDFs.

原文语言：英语

read-python-tokenizer-source

Use this skill to understand the exact Python source code in /root/Tokenizer.py before translating to Scala. This covers every class, method, regex pattern, enum value, default value, and metadata key that must be faithfully reproduced.

原文语言：英语

read-scala-test-spec

Use this skill to read and understand /root/TokenizerSpec.scala to determine the exact expected API signatures, imports, class names, method names, and enum values that the Scala implementation must match.

原文语言：英语

scala-tokenizer-implementation-guide

Use this skill when writing /root/Tokenizer.scala. It provides the concrete implementation strategy, Scala idioms, and compilation workflow for translating the Python Tokenizer to Scala 2.13.

原文语言：英语

email-json-parsing-and-meeting-extraction

Use this skill when you need to parse meeting schedule request emails from a JSON file, extract meeting duration, date/time constraints, sender information, and message IDs. Handles various natural language patterns for expressing meeting preferences.

原文语言：英语

pdf-calendar-parsing

Use this skill when you need to extract calendar/schedule information from a PDF file, including measuring visual positions of appointment blocks, identifying colors of blocks, and determining time slots based on grid lines. Specifically useful for parsing…

原文语言：英语

glm-observation-merge-rmse

How to merge GLM simulation output with field observations and compute RMSE metrics for Lake Mendota calibration. Use this when computing overall RMSE, annual deep RMSE, and summer deep RMSE from matched observation-simulation pairs.

原文语言：英语

glm-output-extraction

How to extract water temperature profiles from GLM NetCDF output files (output.nc). Use this skill when you need to read GLM simulation results, extract depth and temperature arrays, and construct a dataframe of simulated temperatures at specific depths and…

原文语言：英语

database-search-ohio-trip

Use this skill to understand the database file structure and how to search for Ohio cities, restaurants, accommodations, attractions, and distances in the provided dataset files for trip planning.

原文语言：英语

ohio-trip-itinerary-planning

Use this skill when planning a 7-day road trip itinerary from Minneapolis to three cities in Ohio, covering budget, pet-friendly accommodations, cuisine preferences, and no-flight constraints. Produces a JSON itinerary file.

原文语言：英语

convert-keyframes-to-grayscale

Use this skill to convert extracted keyframe PNG images to grayscale in-place using Python3 and PIL/Pillow. Only convert keyframe images, NOT template images (coin.png, enemy.png, turtle.png).

原文语言：英语

count-objects-in-frames-using-template-matching

Use this skill to count coins, enemies, and turtles in grayscale keyframe images using OpenCV template matching with a provided count_objects.py script. Includes robust output parsing and fallback to direct template matching if the script is unavailable.

原文语言：英语

generate-counting-csv

Use this skill to generate the final CSV file at /root/counting_results.csv with columns frame_id, coins, enemies, turtles. Frame IDs must be in the format /root/keyframes_%03d.png.

原文语言：英语

video-keyframe-extraction

Use this skill to extract keyframes (I-frames) from an MP4 video file using ffmpeg. This produces numbered PNG files in a specified output directory.

原文语言：英语

excel-lookup-formulas-gnumeric-compatible

How to write lookup formulas (INDEX/MATCH, VLOOKUP, SUMPRODUCT-based lookups) that work in both Excel and Gnumeric/LibreOffice for multi-condition lookups. Use this when filling cells with formulas that reference data based on multiple criteria.

原文语言：英语

gdp-weighted-mean-net-exports-gcc

How to calculate net exports as % of GDP, descriptive statistics, and weighted mean for GCC countries in the gdp.xlsx workbook. Use this for the specific task of filling in the Task sheet with lookup formulas, percentage calculations, and SUMPRODUCT weighted…

原文语言：英语

inspect-excel-structure

How to inspect the structure of an Excel workbook before writing formulas, including checking sheet names, cell values, data types, and layout. Use this whenever you need to understand the actual content of an Excel file before modifying it.

原文语言：英语

anthropic-brand-tokens

Provides the official Anthropic brand color palette and typography tokens. Use this skill whenever you need canonical HEX values for Anthropic brand colors or the correct heading font name for any Anthropic design artifact.

原文语言：英语

qilv-tonal-patterns

Use this skill to look up the four canonical tonal patterns for seven-character regulated verse (七律 Qilv). Provides exact 平/仄 sequences for all 8 lines under each of the four standard forms, plus the rule for checking positions 2, 4, 6.

原文语言：英语

fill-sc100-small-claims-form

Use this skill to fill in the California SC-100 Small Claims Court PDF form with case data and save the filled PDF. Handles text fields, checkboxes, and radio buttons using pypdf. Run the inspection skill first to confirm field names, then use this skill to…

原文语言：英语

inspect-pdf-fields

Use this skill to inspect all form fields in a PDF file, printing their names, types, and current values/export values. Useful for understanding the structure of a PDF form before filling it.

原文语言：英语

dbscan-custom-metric-clustering

How to run DBSCAN with a custom anisotropic distance metric using sklearn, including defining callable metrics and computing cluster centroids from labeled output.

原文语言：英语

greedy-centroid-matching

How to perform greedy nearest-neighbor matching between two sets of 2D points (cluster centroids vs expert annotations) with a maximum distance threshold, returning matched pairs for F1/delta computation.

原文语言：英语

pareto-frontier-computation

How to compute the Pareto frontier from a set of (F1, delta) optimization results where F1 is maximized and delta is minimized. Includes filtering, dominance checking, and CSV output formatting.

原文语言：英语

find-trivy-and-cache

Use this skill first to locate the trivy binary and its pre-downloaded vulnerability database cache on the system before running any security audit.

原文语言：英语

run-trivy-audit

Use this skill to run Trivy in offline mode against /root/package-lock.json and produce the security audit CSV at /root/security_audit.csv. Use only after locating the trivy cache directory.

原文语言：英语

find-earthquake-farthest-from-pacific-boundary

Find the earthquake within the Pacific plate (Code == "PA") that is farthest from the Pacific plate boundary lines in PB2002_boundaries.json. Uses an equal-area projection for accurate distance measurement. Filters boundaries where PlateA or PlateB equals…

原文语言：英语

load-and-inspect-geodata

Load GeoJSON files with GeoPandas and inspect their structure — columns, CRS, geometry types, and sample rows. Use this before any spatial analysis to understand field names and data layout.

原文语言：英语

count-renaissance-stocks

Count the number of distinct stocks (equity positions only) held by Renaissance Technologies. Filters to SH (share) type only and excludes options/derivatives by checking PUTCALL column. Use this specifically for Q2 answer about Renaissance stock count.

原文语言：英语

get-fund-holdings

Load the INFOTABLE (holdings) parquet for a given quarter and filter by accession_number to get a specific fund's holdings. Handles column name variations. Use this after obtaining an accession_number from the coverpage.

原文语言：英语

load-13f-coverpage

Load and search the COVERPAGE parquet file for a specific fund manager using fuzzy matching on the FILINGMANAGER_NAME field. Returns the best matching row including accession_number, AUM, and other fund details. Use this when you need to find a fund's…

原文语言：英语

github-search-issues

Use `gh search issues` to fetch issues from a specific repo within a date range. Returns structured JSON. More reliable than `gh issue list --search` for date-range filtering.

原文语言：英语

github-search-prs

Use `gh search prs` to fetch pull requests from a specific repo within a date range. Returns structured JSON for downstream processing. More reliable than `gh pr list --search` for date-range filtering.

原文语言：英语

write-report-json

Compile PR and issue metrics into a report.json file at /app/report.json using the required schema for the December community pulse write-up.

原文语言：英语

check-and-setup-python310

Use this skill when you need to find, install, or configure a Python 3.10 environment on the system. Checks for existing python3.10 binary, installs if missing, and sets up pip for that version.

原文语言：英语

implement-simpo-loss

Use this skill to implement the simpo_loss function in SimPOTrainer based on the SimPO paper. The loss combines a length-normalized reward with a margin gamma and uses BCE loss without a reference model.

原文语言：英语

install-simpo-dependencies-python310

Use this skill to install all required packages for the SimPO project using Python 3.10 specifically. Installs torch, transformers, trl, and other dependencies into the python3.10 environment.

原文语言：英语

run-simpo-with-python310-and-log

Use this skill to execute the SimPO unit test with Python 3.10, save the loss results to /root/loss.npz, and log Python version and package info to /root/python_info.txt using the correct Python 3.10 executable.

原文语言：英语

fill-docx-template

Use this skill to fill placeholders in a .docx template file with data from a JSON file, handle conditional sections, and save the result. Covers body paragraphs, tables, and headers/footers.

原文语言：英语