How does it save tokens?

It consolidates multiple tool calls and responses into a single message, reducing the repetitive prompt overhead and metadata tokens typically generated by sequential read operations.

Can I limit the output for very large files?

Yes, you can use the 'context_lines' parameter to specify the number of lines to read per file, preventing the context window from being overwhelmed by large files.

What is the main benefit of the Grep and Read skill?

The main benefit is efficiency; it reduces the time spent exploring codebases by 50-70% by eliminating the need for multiple 'Grep' then 'Read' round-trips.

Does it support regular expressions?

Yes, it leverages the standard Grep tool which supports robust regex patterns for highly targeted searching.

How many files can this skill read at once?

By default, the skill reads the first 5 matching files, but this limit can be adjusted using the 'max_files' parameter to suit your needs.

Grep and Read

Name: Grep and Read
Author: cowwoc

bycowwoc

•

생산성 및 워크플로

Accelerates codebase exploration by combining file searching and reading into a single, high-performance operation.

소개

Grep and Read is an efficiency-focused skill for Claude Code designed to eliminate the latency associated with sequential file discovery and inspection. By coordinating the search and retrieval process, this skill allows Claude to find a pattern and read all relevant files in a single round-trip, resulting in 50-70% faster exploration and significant token savings. It is specifically optimized for developers investigating cross-file bugs, researching implementation patterns, or analyzing documentation across large repositories.

주요 기능

2 GitHub stars
Up to 70% reduction in interaction latency
Regex-supported pattern matching with glob filtering
Configurable context windows and file count limits
Parallelized file reading for multi-match search results
Significant token savings (4,000-8,000 tokens per search)

사용 사례

Investigating error handling logic spread across multiple source files
Reviewing documentation and architectural guides for specific keyword mentions
Exploring API implementations and their associated unit tests simultaneously

소개

주요 기능

2 GitHub stars
Up to 70% reduction in interaction latency
Regex-supported pattern matching with glob filtering
Configurable context windows and file count limits
Parallelized file reading for multi-match search results
Significant token savings (4,000-8,000 tokens per search)

사용 사례

Investigating error handling logic spread across multiple source files
Reviewing documentation and architectural guides for specific keyword mentions
Exploring API implementations and their associated unit tests simultaneously