@6171/json-utils

JSON Utils

Core Capabilities

Action	Description	Command
Format	Beautify or minify JSON	`--action format --input file.json`
Validate	Check if JSON is valid	`--action validate --input file.json`
Diff	Deep-compare two JSON files	`--action diff --input a.json --input2 b.json`
Convert	Convert JSON → YAML/TOML/JSONL	`--action convert --input file.json --to yaml`
Query	Extract value by dot-notation path	`--action query --input file.json --query config.database.host`

# Format JSON
python3 skills/json-utils/scripts/json_utils.py --action format --input messy.json

Minify (compact)

python3 skills/json-utils/scripts/json_utils.py --action format --input data.json --compact

Validate

python3 skills/json-utils/scripts/json_utils.py --action validate --input data.json

Diff two files

python3 skills/json-utils/scripts/json_utils.py --action diff --input old.json --input2 new.json

Convert to YAML

python3 skills/json-utils/scripts/json_utils.py --action convert --input data.json --to yaml

Query nested path

python3 skills/json-utils/scripts/json_utils.py --action query --input config.json --query app.database.host

Save output to file

python3 skills/json-utils/scripts/json_utils.py --action format --input messy.json --output clean.json

JSON output (for piping)

python3 skills/json-utils/scripts/json_utils.py --action validate --input data.json --json


## Examples

### Format
```bash
# Before: {"name":"tea","scores":[1,2,3]}
# After:
# {
#   "name": "tea",
#   "scores": [1, 2, 3]
# }

$ python3 scripts/json_utils.py --action diff --input v1.json --input2 v2.json
📊 Diff: v1.json ↔ v2.json
   Added: 2
   Removed: 1
   Changed: 3

$ python3 scripts/json_utils.py --action query --input config.json --query database.host
📌 database.host = localhost  (str)

Convert to YAML requires pyyaml (pip install pyyaml)
Convert to TOML requires tomli-w (pip install tomli-w)
JSONL convert requires a top-level JSON array; each item becomes one line
Diff returns first 20 items for each change type (added/removed/changed) — large files get truncated
Query uses simple dot-notation only (no array wildcards like [*]). For array access use items.0.name