Stump FAQs

Question 1

What is Stump?

Accepted Answer

Stump is a high-performance CLI and MCP server tool designed to generate compact, token-efficient directory tree visualizations. It's specifically optimized for consumption by large language models (LLMs) like Anthropic's Claude.

Question 2

What are Stump's output options for large directories?

Accepted Answer

For smaller trees, Stump outputs directly to stdout with configurable token limits. For very large directories, it supports writing the tree visualization to a specified file, bypassing stdout token restrictions.

Question 3

How does Stump achieve token efficiency?

Accepted Answer

Stump utilizes a compact JSON output format, resulting in a 50%+ reduction in token usage compared to traditional tree visualization tools. This makes it ideal for working within LLM context windows.

Question 4

Can I customize the directory tree output?

Accepted Answer

Absolutely. Stump offers extensive configuration, including setting traversal depth, smart filtering by file extensions or patterns, showing/hiding hidden files, and toggling performance metrics. It also handles symlinks and large directories with safeguards.

Question 5

What is the MCP server integration?

Accepted Answer

Stump functions as an MCP (Multi-Modal Compute Protocol) server tool, enabling AI agents (like Claude) to invoke it to obtain structured, token-optimized directory information directly, enhancing their ability to understand and interact with codebases.

Stump

Stump

主要功能

使用案例

主要功能

使用案例