Largefile MCP Server
Navigate, search, and edit large codebases, logs, and data files that exceed AI context limits.
Why Largefile?
- Go beyond context limits - Read, search, and edit files too large to fit in AI context windows
- Semantic code navigation - Tree-sitter extracts functions/classes for Python, JS/TS, Rust, Go
- Fewer LLM errors - Search/replace editing eliminates line number mistakes common with line-based edits
- Smart search - Fuzzy matching, regex, case-insensitive, inverted, and count-only modes
- No size limits - Handles multi-GB files via tiered memory strategy (RAM → mmap → streaming)
Quick Start
Prerequisite: Install uv for the uvx command.
{
"mcpServers": {
"largefile": {
"command": "uvx",
"args": ["--from", "largefile", "largefile-mcp"]
}
}
}
Tools
| Tool | Use For |
|---|---|
get_overview |
File structure and semantic outline before diving in |
search_content |
Finding patterns, counting occurrences, regex matching |
read_content |
Reading specific sections; tail/head modes for logs |
edit_content |
Safe search/replace with automatic backups |
revert_edit |
Recovering from bad edits |
list_directory |
Browse directory trees with recursive depth control |
search_directory |
Search patterns across all files in a directory |
When to Use Largefile
Use when:
- File exceeds ~1000 lines or 100KB (supports multi-GB files)
- Navigating large codebases with semantic structure
- Analyzing log files (especially recent entries with tail mode)
- Making search/replace edits across large files
- Counting occurrences without loading full content
Don't use for:
- Small files that fit in context (AI doesn't need help with those)
- Binary files (images, executables, compressed)
Usage Examples
Large Codebase Navigation
# Get semantic structure of a large Python file
overview = get_overview("/path/to/large_module.py")
# Returns: 2,847 lines, 15 classes, function outline via Tree-sitter
# Find all class definitions
classes = search_content("/path/to/large_module.py", "class ", fuzzy=False)
# Read complete class with semantic chunking
code = read_content("/path/to/large_module.py", pattern="class UserModel", mode="semantic")
Batch Refactoring
# Preview rename across file
preview = edit_content("/path/to/api.py", changes=[
{"search": "process_data", "replace": "transform_data"},
{"search": "old_endpoint", "replace": "new_endpoint"}
], preview=True)
# Apply changes (creates automatic backup)
result = edit_content("/path/to/api.py", changes=[...], preview=False)
# Undo if needed
revert_edit("/path/to/api.py")
Log Analysis
# Get log file overview
overview = get_overview("/var/log/app.log")
# Returns: 150,000 lines, 2.1GB
# Read last 500 lines efficiently
recent = read_content("/var/log/app.log", limit=500, mode="tail")
# Count errors without loading content
error_count = search_content("/var/log/app.log", "ERROR", count_only=True, fuzzy=False)
# Find errors with regex
errors = search_content("/var/log/app.log", r"ERROR.*timeout", regex=True)
Supported Languages
Tree-sitter semantic analysis for: Python, JavaScript/JSX, TypeScript/TSX, Rust, Go, Java
Other file types use text-based analysis with graceful fallback.
File Size Handling
| Size | Strategy |
|---|---|
| < 50MB | Full memory loading with AST caching |
| 50-500MB | Memory-mapped access |
| > 500MB | Streaming (tail/head modes recommended) |
Configuration
Environment variables for tuning:
LARGEFILE_MEMORY_THRESHOLD_MB=50 # RAM loading limit
LARGEFILE_MMAP_THRESHOLD_MB=500 # Memory mapping limit
LARGEFILE_FUZZY_THRESHOLD=0.8 # Match sensitivity (0.0-1.0)
LARGEFILE_MAX_SEARCH_RESULTS=20 # Results per search
LARGEFILE_BACKUP_DIR=~/.largefile/backups
Documentation
- API Reference - Detailed tool documentation
- Configuration Guide - All environment variables
- Examples - More workflow examples
- Design Document - Architecture details
- Contributing - Development setup
Metadata
Release files for largefile 0.3.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| largefile-0.3.0.tar.gz | 2.6 MB | Without provenance | Details for largefile-0.3.0.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| largefile-0.3.0-py3-none-any.whl | Python 3 | none | any | Details for largefile-0.3.0-py3-none-any.whl |
Total release size: 2.7 MB
Release files / largefile-0.3.0.tar.gz
| Download URL | largefile-0.3.0.tar.gz |
|---|---|
| Size | 2.6 MB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
31f8d9d71021e6e3fcb588a5c7b2daf891dcf522c27c39ab4d893f0693f8044a
|
|
BLAKE2b-256 checksum How to use checksums |
f4d12f44c61b5127c29fc25ea76c06f9aeff172b5a9129f6db5f424965ab56b2
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
uv/0.11.5 {"installer":{"name":"uv","version":"0.11.5","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"Ubuntu","version":"24.04","id":"noble","libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":true}
|
Release files / largefile-0.3.0-py3-none-any.whl
| Download URL | largefile-0.3.0-py3-none-any.whl |
|---|---|
| Size | 39.3 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
94596038b04339c8ec2d98e5ece28635deed1880e8bedf429aa241e8801e46ac
|
|
BLAKE2b-256 checksum How to use checksums |
52948228f5e9d68677921ceeeffede0dba4ef5b0ea104be2878fed932d76d7b8
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
uv/0.11.5 {"installer":{"name":"uv","version":"0.11.5","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"Ubuntu","version":"24.04","id":"noble","libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":true}
|
Security for largefile 0.3.0
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.