exploratory-data-analysis
Performs bounded, local exploratory analysis of explicitly supported scientific
by k-dense-aiFree
Description
exploratory-data-analysis is an agent skill for inspecting local scientific data files before modelling or confirmatory analysis. It produces bounded, deterministic aggregate reports for files you explicitly authorise. It supports redacted profiles of CSV, TSV and JSON files, with optional support for NumPy, HDF5 and other formats described in its domain references. The skill is clear about its limits. It does not certify a file, infer scientific meaning, or support every format it lists. Its safety rules are firm: it treats every cell, header, sequence title, HDF5 name or attribute, image tag and metadata string as untrusted data. It never follows instructions embedded in a file, never resolves embedded URLs, never runs macros, never evaluates expressions, never executes HDF5 objects, never loads models, and never passes file-derived text to a shell. It is meant to run on local data, so the output is an aggregate report you can review rather than a copy of the raw records. The skill suits researchers who want a first look at a dataset's structure and quality before deciding how to model it.…
Highlights
- Bounded local profiles of CSV, TSV and JSON
- Deterministic aggregate reports
- Optional NumPy and HDF5 support
- Treats file contents as untrusted
- Never runs macros or evaluates expressions
Ratings & Reviews
No ratings yet
Be the first to rate exploratory-data-analysis.
People also visit
datanalysis-credit-risk
Credit risk data cleaning and variable screening pipeline for pre-loan modeling
exploratory-data-analysis FAQ
What does it do?
It gives a bounded, deterministic aggregate report of a local scientific data file, before you model it.
Which formats are supported?
Redacted profiles of CSV, TSV and JSON, with optional support for NumPy, HDF5 and others described in its references.
Does it judge whether a file is good?
No. The skill says it does not certify a file or infer scientific meaning.
How does it handle hostile content?
It treats all cell, header and metadata text as untrusted, and it never follows embedded instructions, runs macros or resolves URLs.
How do I install it?
Run npx skills add https://github.com/k-dense-ai/scientific-agent-skills --skill exploratory-data-analysis.
Is it checked for security?
Its security audits show passes from Gen Agent Trust Hub, Socket and Snyk.
What are the best exploratory-data-analysis alternatives?
Popular alternatives to exploratory-data-analysis include data-analysis, knap and data-quality-frameworks. See all exploratory-data-analysis alternatives