chunked-file-reading

Locate keyword matches in large files and read focused line windows.

1|Updated Nov 25, 2025

Search Tags:#jsonl #large-files #log-reading #offset-limit #chunked-reading #window-reading

Ready to run

npx skills add https://github.com/89jobrien/dotfiles --skill chunked-file-reading

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: chunked-file-reading
Source: https://github.com/89jobrien/dotfiles/tree/main/dot-claude/skills/chunked-file-reading
Command: npx skills add https://github.com/89jobrien/dotfiles --skill chunked-file-reading

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Large files such as logs, JSON Lines, and CSVs can exceed context limits or slow down analysis, forcing you to read and search through huge data without precise targeting.

Core Features & Use Cases

Windowed reads: read only a small window around a match using offset and limit to minimize data transfer and processing.
Grep-based localization: quickly locate relevant sections in multi-thousand-line files before reading.
JSONL/line-based navigation: navigate across lines to locate and inspect specific entries without loading the entire file.
Use Case: debugging a failing service with a massive log to locate the error line and read surrounding lines for context.

Quick Start

Grep for a keyword to locate the section, then read a focused window around the match.

Frequently Asked Questions about chunked-file-reading

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I read large log files without exceeding context limits?▼

Reading large log files without exceeding context limits requires identifying a keyword to locate the relevant line range, then reading a focused window around the match using offset and limit. This minimizes data transfer and preserves surrounding context.

What is the best way to navigate massive JSONL files to inspect specific entries?▼

The best way to navigate massive JSONL files is using line-based navigation. You locate relevant sections by searching for a keyword, then iteratively move the window forward or backward by adjusting the offset to inspect specific entries without loading the entire file.

How does windowed reading work for large CSVs?▼

Windowed reading for large CSVs works by reading only a small, focused data window around a keyword match. You use offset and limit parameters to define this window, minimizing processing overhead and avoiding context limit issues associated with huge data.

Can I search for an error in a multi-thousand-line file and read surrounding lines for context?▼

Yes, you can search for an error in a multi-thousand-line file and read surrounding lines for context. The process uses grep-based localization to find the error line, followed by a windowed read to examine the surrounding lines without loading the entire log.

Does chunked file reading support moving backward through a file?▼

Yes, chunked file reading supports moving backward through a file. After locating a section via keyword, you can iteratively move the reading window forward or backward by adjusting the offset parameter to navigate the file as needed.