Ships three tools for RFC 4180 compliant CSV handling: parse_csv, to_csv, and pluck_columns. It correctly deals with quoted fields, embedded commas, byte-order marks, and CRLF line endings instead of letting the LLM hallucinate CSV parsing logic. Part of a fourteen-server collection aimed at giving agents reliable primitives for transforms they routinely botch when left to generate code. Install it via npx in Claude Desktop, Cursor, Cline, or any MCP client when you're working with real-world CSV data that breaks naive split-on-comma approaches. Pure JavaScript implementation, no external dependencies or Rust toolchain required.
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.
| Package | Tools | When to install |
|---|---|---|
@mukundakatta/promptbudget-mcp | truncate_to_token_budget | LLM needs to fit text into a context window mid-task |
@mukundakatta/citecite-mcp | inject_citations, parse_citations, strip_citations | Round-trip [1] [2] markers in RAG outputs |
@mukundakatta/ragmetric-mcp | recall_at_k, hit_at_k, mrr, ndcg_at_k, evaluate_batch | LLM helps score retrieval quality |
@mukundakatta/ragdrift-mcp | interpret_drift_score, recommend_thresholds, explain_drift_dimensions | LLM diagnoses RAG drift alerts |
| Package | Tools | When to install |
|---|---|---|
@mukundakatta/csv-tools-mcp | parse_csv, to_csv, pluck_columns | LLM needs RFC 4180-correct CSV (quoted fields, embedded commas, BOMs) |
@mukundakatta/regex-test-mcp | test_regex, find_all, replace | LLM needs exact regex semantics with real match offsets and named groups |
@mukundakatta/jmespath-mcp | json_query | LLM needs deep JSON traversal (filters, projections, pipes) without hallucinating |
@mukundakatta/diff-mcp | unified_diff, apply_patch, parse_patch | Code-review or code-edit agent needs character-precise patches |
@mukundakatta/sqlfmt-mcp | format_sql, list_dialects | Deterministic SQL formatting across 19 dialects |
@mukundakatta/shellquote-mcp | quote_bash, quote_bash_argv, quote_cmd, quote_powershell | Safe shell argument escaping; LLMs constantly mishandle quotes/$VAR |
@mukundakatta/json5-mcp | parse_json5, to_json5, to_strict_json | LLM emitted JSON-with-comments / trailing-commas; round-trip to strict JSON |
@mukundakatta/toml-yaml-json-mcp | parse, format, convert | TOML / YAML / JSON conversion (LLMs especially mishandle TOML) |
@mukundakatta/timezone-mcp | convert_tz, now_in, tz_offset | IANA timezone math with real DST rules; LLMs hallucinate offsets |
@mukundakatta/html-to-markdown-mcp | html_to_md, extract_text | Web-scraping agents need clean Markdown / text from HTML |
The first four wrap or re-implement logic from the user's Rust + Python work:
ragdrift (crates.io) /
ragdrift-py (PyPI) — five-dimensional drift detectionembedrank, promptbudget,
stopstream, citecite,
ragmetric — pure-Rust crates from
the rust-llm-stack workspaceThe MCP servers are independent: they implement the logic they need in plain JavaScript so installing them doesn't pull a Rust toolchain.
Add to your MCP client config. Example for Claude Desktop:
{
"mcpServers": {
"promptbudget": { "command": "npx", "args": ["-y", "@mukundakatta/promptbudget-mcp"] },
"citecite": { "command": "npx", "args": ["-y", "@mukundakatta/citecite-mcp"] },
"ragmetric": { "command": "npx", "args": ["-y", "@mukundakatta/ragmetric-mcp"] },
"ragdrift": { "command": "npx", "args": ["-y", "@mukundakatta/ragdrift-mcp"] },
"csv-tools": { "command": "npx", "args": ["-y", "@mukundakatta/csv-tools-mcp"] },
"regex-test": { "command": "npx", "args": ["-y", "@mukundakatta/regex-test-mcp"] },
"jmespath": { "command": "npx", "args": ["-y", "@mukundakatta/jmespath-mcp"] },
"diff": { "command": "npx", "args": ["-y", "@mukundakatta/diff-mcp"] },
"sqlfmt": { "command": "npx", "args": ["-y", "@mukundakatta/sqlfmt-mcp"] },
"shellquote": { "command": "npx", "args": ["-y", "@mukundakatta/shellquote-mcp"] },
"json5": { "command": "npx", "args": ["-y", "@mukundakatta/json5-mcp"] },
"toml-yaml-json": { "command": "npx", "args": ["-y", "@mukundakatta/toml-yaml-json-mcp"] },
"timezone": { "command": "npx", "args": ["-y", "@mukundakatta/timezone-mcp"] },
"html-to-md": { "command": "npx", "args": ["-y", "@mukundakatta/html-to-markdown-mcp"] }
}
}
git clone https://github.com/MukundaKatta/mcp-stack
cd mcp-stack
npm install
npm test --workspaces
MIT OR Apache-2.0.
com.mcparmory/google-sheets
domdomegg/google-sheets-mcp
henilcalagiya/google-sheets-mcp
cct15/war-dashboard-data
moooonad/mcp-google-sheets-full
io.github.br0ski777/csv-to-json