How to use
- Copy the installation command and review the source package before installing it in your own agent environment.
- Check the source document for required tools, credentials and supported hosts.
- Describe your input and desired deliverable; review the output against the source material.
npx skills add https://github.com/xsavikx/okf-skills --skill okf-csvBefore you start
- Installation does not include a model subscription, API allowance or permission to access your files.
- Repository stars and installation counts describe activity, not an independent quality or security test.
Rendered from the source repository. Headings, examples, code, tables, links, and referenced images are preserved.
CSV OKF Connector
This skill provides a Go-based CLI tool to document a directory of CSV files as an Open Knowledge Format (OKF) bundle: one concept per CSV, with an inferred column schema and an optional data profile, and to sync enriched descriptions back to a .okf-metadata.yaml sidecar.
When to Use
Use this skill when you need to:
- Catalog a folder of CSV files — each becomes a
CSV Fileconcept undertables/,
with a # Columns table whose types are inferred by sampling values.
- Capture per-column statistics (
--profile) and sample rows (--sample) as
grounding for enrichment.
- Round-trip enriched descriptions back to the source via
.okf-metadata.yaml.
CSV has no declared types or comment store, so column types are inferred and descriptions live in the sidecar (the same pattern okf-fs uses).
Setup
# Install the published binary (Go 1.24+):
go install github.com/xSAVIKx/okf-skills/skills/okf-csv@latest
# …or build from a clone:
cd skills/okf-csv && go build -o okf-csv .How to Use
1. Produce an OKF bundle
./okf-csv produce --dir <csv-directory> --out <bundle-dir> [--sample <N>] [--profile]--dir(required): directory of CSV files (traversed recursively, honoring.okfignore).--out(required): output OKF bundle directory.--sample <N>: embed up to N sample rows per file as a## Samplesection.--profile: compute per-column statistics (non-null, null, distinct, min, max,
detected semantic type, low-cardinality value sets) into a ## Data Profile section.
Each data/orders.csv becomes tables/orders.md. Re-running preserves unchanged concepts byte-for-byte (incremental produce), keeping any enriched descriptions.
2. Ingest / sync descriptions
./okf-csv ingest --dir <csv-directory> --bundle <bundle-dir> [--sync]- Verifies each concept's columns still match its CSV header (reports drift).
--sync: writes the bundle's descriptions back to.okf-metadata.yamlin--dir.
3. Inspect the schema (self-description)
./okf-csv schemaPrints the machine-readable JSON description used by okf-mcp to expose the skill.
