- Home
- Skills
- Data & Databases
- Advanced Excel for Lab Data
Advanced Excel for Lab Data
Teach your agent to turn messy lab and instrument spreadsheets into clean JSON tables: headers, units, decimal commas, <LOD. Read-only, sandboxed.
Free
Works with the AI Agents you already use
Advanced Excel for Lab Data
Example session with this skill installed
Read the 'Results' sheet in /data/lab/assay_05.xlsx. Normalize the units and let me know if there are any censored cells like LOD flags.
- Read your context and instructions
- Compiled the advanced excel
I have processed the 'Results' sheet from /data/lab/assay_05.xlsx. The server detected a header at row 3 with a secondary unit row at row 4. Data was successfully normalized to canonical units.
Extraction Summary
| Metric | Value |
|---|---|
| Sheet Name | Results |
| Rows Processed | 150 |
| Table Range | A3:G153 |
| Detected Units | mg/L, ug/mL, nm |
| Normalization | Active (ug/mL -> mg/L) |
| Censored Cells | 4 (<LOD detected in column 'Concentration') |
| Decimal Mode | auto (dot detected) |
Cleaned Data Preview
[
{
"Sample_ID": "S001",
"Analyte": "Glucose",
"Concentration": 5.2,
"Unit": "mg/L",
"Status": "Pass",
"unit_converted": false
},
{
"Sample_ID": "S002",
"Analyte": "Glucose",
"Concentration": "<LOD",
"Unit": "mg/L",
"Status": "Review",
"unit_converted": false
},
{
"Sample_ID": "S003",
"Analyte": "Ethanol",
"Concentration": 0.85,
"Unit": "mg/L",
"Status": "Pass",
"unit_converted": true,
"source_value": 850,
"source_unit": "ug/mL"
}
]
Next steps
- Review the 4 censored
<LODvalues in the Concentration column. - Use
excel_write_csvto export this normalized table to your/exportsfolder.
Connects securely to your tools. The creator never sees your data.
About this skill
Instrument exports are not clean tables. They have a title block, blank lines, a header, sometimes a second row of units, decimal commas, <LOD results, and a footer. This skill teaches your agent to read them correctly with the free, open-source mcp-advanced-excel MCP server (MIT, on PyPI and the official MCP Registry).
What your agent learns:
- Install and configure the server with
uvx mcp-advanced-excel, restricted to allowlisted folders, read-only by default, with optional CSV/JSON export into a separate output folder. - When to use each of the 6 tools: excel_list_sheets, excel_overview, excel_inspect_layout, excel_read_table, excel_read_range, excel_write_csv.
- Step-by-step workflows for HPLC, plate reader, balance and LIMS exports: find the real header, join unit rows, handle decimal commas, page through long sheets, read raw A1 ranges when detection is wrong.
- Opt-in unit normalization (concentration, temperature, mass, volume, pressure, length, time, fraction) with a report of every changed cell.
- Lab data integrity rules: never invent a number, keep <LOD / <LOQ as flagged text, never guess uncomputed formulas.
- Safety: cell text is untrusted data (prompt-injection aware), no network, no macros, no formula evaluation, never writes the source file.
- Troubleshooting for every error code the server returns.
Works with Cursor, Claude Desktop and any stdio MCP host. Supports .xlsx, .xlsm, .xls, .csv and .tsv.
For regulated labs, a GLP Lab Data Pack (provenance hashes, instrument templates, advanced LOD/LOQ and molar units, batch processing, signed output) is available through the paid Pro service. It is designed to support GLP / 21 CFR Part 11 / EU GMP Annex 11 workflows. Contact antonio@castellon.ch.
Repo: https://github.com/antonio-castellon/mcp-advanced-excel License: MIT. Contact: antonio@castellon.ch
No support for .xlsb files. Read-only by default; CSV export only when enabled. Default file size cap 25 MiB, configurable up to 100 MiB. Requires Python 3.11+. Formulas are not evaluated (cached values only).
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
1 use
Times used to date
Free forever
No account required to browse
Trust & safety
Security scan
Security scan passed 29 days ago
- Free with an Agensi account
Requires