Skip to main content

CLI AI Data Analyst — query CSV and Excel files in plain English

Project description

🧠 QueryMind

Ask questions about your data in plain English. No SQL. No code. Just a terminal.

QueryMind is a CLI data analyst that lets you load a CSV or Excel file and query it conversationally — right in your terminal.

>> top 5 regions by sales
>> which month had the highest profit?
>> average spend by payment method in ascending order
>> show sales in sheet Orders by customer segment

Install

pip install querymind-cli

Requirements:

  • Python 3.10+
  • Ollama (optional — enables LLM fallback for complex queries)

If you want LLM support, install Ollama and pull the model:

ollama pull phi

Quickstart

querymind

You'll be prompted to:

  1. Enter a CSV or Excel file path
  2. Select sheets (Excel only)
  3. Map your metric and dimension columns
  4. Start asking questions

What it can do

Query What happens
top 5 products by revenue Ranked bar chart in terminal
which region had lowest sales Ascending comparison with insight
average profit by category Mean aggregation per group
sales trend over time monthly Monthly groupby on datetime column
show sales in sheet Orders by region Sheet-scoped query
which manager had the most sales Cross-sheet join (Orders + Users)
sales by region in ascending order Explicit sort order

Supported file formats

Format Extension
CSV .csv, .tsv
Excel .xlsx, .xls, .xlsm

Auto-detects: encoding (UTF-8 BOM, latin-1), delimiter (comma, semicolon, tab, pipe), packed integer dates (DDMMYYYY, YYYYMMDD).


How it works

Your query
    ↓
InputGuard       — blocks gibberish and sensitive input
    ↓
InterpreterAgent — rule-based intent extraction (fast, no LLM needed)
    ↓
LLMInterpreter   — Ollama fallback for complex queries (optional)
    ↓
JoinResolver     — auto-detects and performs cross-sheet joins
    ↓
Analyzer         — pandas groupby / aggregation
    ↓
InsightGenerator — formats result + ASCII bar chart

Session History

Every query and answer is automatically saved to:

~/querymind_sessions/querymind_history.md

This is a single persistent Markdown file — all sessions across all datasets append to it. Open it in any editor or view it on GitHub.

Inside the TUI, type /history to see your last 5 queries and the file path.


Beta

QueryMind is in active development. If something breaks or a query gives a wrong answer, please open an issue with:

  • Your query
  • The column names in your file (no need to share actual data)
  • The output you got

This feedback directly shapes what gets fixed next.


License

MIT

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

querymind_cli-0.1.2.tar.gz (37.1 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

querymind_cli-0.1.2-py3-none-any.whl (43.9 kB view details)

Uploaded Python 3

File details

Details for the file querymind_cli-0.1.2.tar.gz.

File metadata

  • Download URL: querymind_cli-0.1.2.tar.gz
  • Upload date:
  • Size: 37.1 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.13.5

File hashes

Hashes for querymind_cli-0.1.2.tar.gz
Algorithm Hash digest
SHA256 2ef8596fb2e008317f327be359a7041e0eb728d011fcb20ef62ec9ac299614f5
MD5 0c11b7ee40f1333a47ebbf66a5ddb5da
BLAKE2b-256 a87304a02ff4124cd23c8f86ed3cb0ea3fee20a8616a5874dec7781e7baddfff

See more details on using hashes here.

File details

Details for the file querymind_cli-0.1.2-py3-none-any.whl.

File metadata

  • Download URL: querymind_cli-0.1.2-py3-none-any.whl
  • Upload date:
  • Size: 43.9 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.13.5

File hashes

Hashes for querymind_cli-0.1.2-py3-none-any.whl
Algorithm Hash digest
SHA256 7ce5a92e83df6ca794aec20ecc23fcfa215053d64bc8af825c0515c2326166bd
MD5 3d0372d25020f67437dd450c7ab42eb1
BLAKE2b-256 54ad05bfe35d86d55a3ac639bdbbcabcebe5d60f4fc92d69edfd12b0026d478f

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page