MCP.Market

Diff MCP server

by MukundaKatta·io.github.MukundaKatta/diff-mcp·v0.1.0

Character-precise unified diff + patch application + patch parsing.

A92/100grade A
What users say
No reviews yet
Be the first
Safety scan
A92/100

full report

Adoption
New

0 stars

Reviews

Write one

Nobody has reviewed Diff yet.

If you have run it, two minutes of your experience saves the next person an afternoon.

Diff tools (3, 2 write)

write = sends, deletes, buys or posts

Read from the package source without running it. The installed server may list more.

  • apply_patchwrite action

    Apply a unified-diff patch to an original string. Returns success/false plus the patched text. Returns success:false with a reason when the patch can't apply cleanly (e.g. context drift).

  • parse_patchwrite action

    Parse a unified-diff patch and return its structure (file pairs, hunk count, addition/deletion counts per file). Useful for summarizing a patch before deciding whether to apply it.

  • unified_diff

    Generate a unified diff between two strings. Returns the patch text plus counts of additions / deletions / unchanged lines. Use this when an LLM needs to produce a code review or a patch the user (or a tool) will then apply.

Public scan report

scanner v0.1.9 · 2026-09-27 · same rubric, same numbers if you re-run it

no findings
  • Code scan3 source files scanned25/25
  • –Live reliabilityno gateway calls yet and no remote to proben/a
  • –Tool poisoningtools not inspected (local package is not executed); not countedn/a
  • Auth qualitylocal package, no credentials required12/15
  • Maintenancelast push 29 days ago15/15
  • Maintainer identityregistry namespace matches repository owner; GitHub account older than a year8/10
Overall 92/100. Components that don't apply are left out of the denominator. Any critical finding is an F.RubricAppeal a findingJSON

How Diff compares

Against every graded MCP server, from our own scans and uptime checks. Updated after each scan run.

Safety score
92/100
higher than 90% of 48,273 servers
Tools
3
the typical server has 7
Tools that change things
67%
typical: 0%. Fewer write tools means less an agent can break

What its tools let an agent do: 1 read · 1 change · 1 run or deploy

Install Diff in Claude Code, Cursor or VS Code

Runs npx -y @mukundakatta/diff-mcp on your machine. Read the scan report first; the gateway never runs local packages.

claude mcp add diff-mcp -- npx -y @mukundakatta/diff-mcp
Add to Cursor

What the publisher says

From the Diff repository's README, as published. We do not edit it. Read it on GitHub

mcp-stack

Fourteen small MCP servers for everyday LLM / RAG / agent work. Each ships independently on npm, listed in the official MCP Registry, and is callable from Claude Desktop, Cursor, Cline, Windsurf, Zed, and any other MCP-compatible client.

RAG / agent helpers

Reliable transforms LLMs reach for tools instead of imagining

Sibling libraries

The first four wrap or re-implement logic from the user's Rust + Python work:

ragdrift-py (PyPI) — five-dimensional drift detection

  • ragdrift (crates.io) /

stopstream, citecite, ragmetric — pure-Rust crates from the rust-llm-stack workspace

  • embedrank, promptbudget,

The MCP servers are independent: they implement the logic they need in plain JavaScript so installing them doesn't pull a Rust toolchain.

Install (any of them)

Add to your MCP client config. Example for Claude Desktop:

Shortened. The full README is on GitHub.

Nothing above is checked by us. What we check is on the safety report.

Diff: common questions

Is Diff MCP server safe?
Yes, by our scan: it is graded A (92/100). Read the Diff safety report
How do I install Diff?
It runs on your machine. Copy the Claude Code, Cursor, VS Code or Claude Desktop config from the install section.
Does Diff need an API key?
Not as far as the registry entry and our scan can tell: no credentials are declared or required.
Is Diff maintained?
The last commit was 65 days ago (2026-08-29). The latest release is v0.1.0.
How does Diff compare with other MCP servers?
safety score 92/100 (higher than 90% of 48,273 servers); tools 3 (the typical server has 7); tools that change things 67% (typical: 0%. Fewer write tools means less an agent can break). Full comparison.

More from MukundaKatta →