Skip to content

Getting Started

Installation

uv tool install dumpcat

pip install dumpcat

pipx install dumpcat

git clone https://github.com/Allenfp/dumpcat.git
cd dumpcat
pip install .

Basic usage

Run dumpcat in any directory to dump its file tree and contents as Markdown:

dumpcat

Dump a specific directory:

dumpcat src/

Filtering files

Only include certain extensions:

dumpcat -i .py -i .js

Exclude patterns:

dumpcat -e node_modules -e "*.pyc"

Limit directory depth:

dumpcat -d 2

Output options

Write to a file instead of stdout:

dumpcat -o dump.md

Switch output format:

# Plain text
dumpcat -f plain

# JSON
dumpcat -f json

Prepending prompts

Add context for your LLM at the top of the output:

dumpcat -p "Review this code for bugs and suggest improvements"

Or read from a prompt template file:

dumpcat -p @prompts/review.md

Show stats

Get a summary of file count, lines, and estimated tokens:

dumpcat -s

Sending output to a local LLM

Dumpcat can send your dump directly to a local LLM (Ollama, vLLM, LM Studio) or any OpenAI-compatible endpoint:

dumpcat src/ -i .py --llm ollama -m llama3 -p "Review this code for bugs"

Set up reusable LLM targets:

# Create a starter config
dumpcat init

# Edit ~/.dumpcat/dumpcat_profiles.toml, then:
dumpcat src/ --llm -p "Explain this codebase"

Appending to output files

Use -A to append to an existing file instead of overwriting:

dumpcat src/ -o dump.md
dumpcat tests/ -o dump.md -A

What's next?