Mcp Utils
Wire your coding agent to clean CSVs, run stats, transform images, emit reports, persist outputs, and execute shell steps without leaving the chat.
Overview
io.github.followtz/mcp-utils is a MCP server for the Build phase that gives coding agents CSV, cleaning, stats, image, report, save, and shell data-processing tools.
What is this MCP server?
- CSV read/write and tabular cleaning helpers for import pipelines
- Descriptive statistics and numeric summaries on in-agent datasets
- Image-oriented processing hooks for media-heavy side projects
- Report generation and save-to-disk style persistence flows
- Shell execution surface for scripted data prep alongside agent reasoning
- Tool families documented in the catalog: CSV, cleaning, stats, image, report, save, and shell
What problem does it solve?
You keep context-switching out of the agent to scrub CSVs, run quick stats, or script one-off data jobs on your laptop.
Who is it for?
Indie builders automating imports, ad-hoc analytics, and report dumps while shipping integrations with Claude Code or Cursor.
Skip if: Teams that need governed warehouses, scheduled ETL, or enterprise BI instead of agent-local file tooling.
What do I get? / Deliverables
After you register the server, the agent can run structured data-processing steps in-thread and hand you cleaned tables, summaries, and saved outputs.
- Cleaned or transformed datasets ready for your app or script
- Statistical summaries and generated reports the agent can cite in follow-up code
- Saved output files from in-session processing runs
Recommended MCP Servers
Journey fit
Solo builders hit data wrangling while wiring backends, imports, and agent workflows—this MCP belongs on the Build shelf as integration tooling. It exposes MCP tools that plug into Claude Code, Cursor, and similar agents rather than shipping a standalone app UI.
How it compares
MCP data toolkit for agents, not a hosted analytics platform or a single fixed-purpose Claude skill.
Common Questions / FAQ
Who is io.github.followtz/mcp-utils for?
Solo and indie developers who want CSV cleaning, stats, images, reports, and shell-backed prep callable from Claude Code, Cursor, Codex, or other MCP clients.
When should I use io.github.followtz/mcp-utils?
Use it during Build when you are wiring data imports, normalizing exports, or prototyping analytics pipelines without leaving your agent session.
How do I add io.github.followtz/mcp-utils to my agent?
Install or clone the MCP server package from the registry entry, add it to your client’s MCP server config (stdio or your transport), restart the agent, and confirm the CSV, stats, and shell tools appear in the tool list.