EX

explore-references

This searches reference libraries to analyze feature implementation methods and compiles research summaries on best practices.

Install

mkdir -p .claude/skills/explore-references && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/3359" && unzip -o skill.zip -d .claude/skills/explore-references && rm skill.zip

Installs to .claude/skills/explore-references

Activation

This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.

Research how a feature works in reference libraries
51 charsno explicit “when” trigger
Intermediate

Key capabilities

  • Search reference library source code
  • Compare implementation patterns across libraries
  • Analyze error handling and edge case strategies
  • Extract best practices for API design
  • Compile research findings into structured markdown

How it works

The skill systematically searches specified library directories, compares implementation approaches, and synthesizes findings into a research document.

Inputs & outputs

You give it
Feature or concept to research
You get back
Structured research document with recommendations

When to use explore-references

  • Compare PDF parsing logic across libraries
  • Research API patterns for PDF manipulation
  • Identify edge case handling strategies
  • Document library-specific best practices

About this skill

You are researching how a feature or concept is implemented in the reference libraries to inform our own implementation.

Your Task

  1. Understand the request - Based on the conversation context, identify what feature or topic to research. Ask the user if unclear.
  2. Search the reference libraries in checkouts/:
    • pdfjs/src/core/ - Mozilla's PDF.js (parsing focus)
    • pdf-lib/src/ - pdf-lib (TypeScript API patterns)
    • pdfbox/pdfbox/src/main/java/org/apache/pdfbox/ - Apache PDFBox (comprehensive coverage)
  3. Analyze the implementations - How does each library approach this?
  4. Summarize findings - Write a research document to .agents/scratch/

Research Process

Step 1: Search Each Library

For each reference library, search for relevant code:

  • Class definitions and data structures
  • Key methods and algorithms
  • Error handling approaches
  • Edge cases handled
  • Public API surface

Step 2: Compare Approaches

Analyze the differences:

  • What patterns does each library use?
  • What are the tradeoffs of each approach?
  • Which handles edge cases best?
  • Which has the cleanest API?

Step 3: Extract Insights

Identify what we should learn:

  • Best practices to adopt
  • Pitfalls to avoid
  • Edge cases we must handle
  • API patterns that feel ergonomic

Output Format

Write your findings to .agents/scratch/<topic>-research.md with this structure:

# <Topic> Research

## Summary

Brief overview of findings and recommendations.

## pdf.js Approach

- How it works
- Key files: `path/to/file.js`
- Pros/cons

## pdf-lib Approach

- How it works
- Key files: `path/to/file.ts`
- Pros/cons

## PDFBox Approach

- How it works
- Key files: `path/to/File.java`
- Pros/cons

## Recommendations for @libpdf/core

- What approach to take
- Key considerations
- Edge cases to handle

Guidelines

  • Be thorough - This research informs implementation decisions
  • Include code references - File paths and line numbers help future exploration
  • Note edge cases - What weird PDFs do the libraries handle?
  • Consider our constraints - We target Node, Bun, and browsers equally
  • Focus on insights - Don't just describe; analyze and recommend

Begin

Search the reference libraries for the topic determined from the conversation context and compile your research findings.

When not to use it

  • Researching topics outside of PDF manipulation
  • When reference libraries are unavailable in checkouts/

Prerequisites

Access to reference library source code

Limitations

  • Limited to the provided reference libraries
  • Requires manual topic identification

How it compares

It provides a structured, comparative analysis of existing library implementations rather than a general search.

Compared to similar skills

explore-references side by side with the closest alternatives in the catalog.

SkillInstallsUpdatedSafetyDifficulty
explore-references (this skill)16moNo flagsIntermediate
paper-recommended-questions04moReviewBeginner
qmd262moReviewIntermediate
biorxiv-database79moReviewBeginner

Try saying

Example prompts that trigger this skill in your AI assistant.

You might also like

paper-recommended-questions

scholarclaw

Extracts text from PDF files and generates 3 thought-provoking questions that explore the study in depth. Supports direct PDF file processing and integration with OpenClaw.

00

qmd

tobi

Search personal markdown knowledge bases, notes, meeting transcripts, and documentation using QMD - a local hybrid search engine. Combines BM25 keyword search, vector semantic search, and LLM re-ranking. Use when users ask to search notes, find documents, look up information in their knowledge base, retrieve meeting notes, or search documentation. Triggers on "search markdown files", "search my notes", "find in docs", "look up", "what did I write about", "meeting notes about".

2669

biorxiv-database

lifangda

Efficient database search tool for bioRxiv preprint server. Use this skill when searching for life sciences preprints by keywords, authors, date ranges, or categories, retrieving paper metadata, downloading PDFs, or conducting literature reviews.

780

read-arxiv-paper

karpathy

Use this skill when when asked to read an arxiv paper given an arxiv URL

1656

twitter-reader

daymade

Fetch Twitter/X post content by URL using jina.ai API to bypass JavaScript restrictions. Use when Claude needs to retrieve tweet content including author, timestamp, post text, images, and thread replies. Supports individual posts or batch fetching from x.com or twitter.com URLs.

552

notion-research-documentation

openai

Research across Notion and synthesize into structured documentation; use when gathering info from multiple Notion sources to produce briefs, comparisons, or reports with citations.

551

Search skills

Search the agent skills registry