Agent skill

data-transformation

Use when modifying, converting, or transforming structured data with dasel v3 — in-place mutations, format conversion, batch operations, array manipulation, object construction, and merge patterns across JSON, YAML, TOML, XML, CSV, HCL, INI

View SKILL.md on GitHub Repository
Stars 33
Forks 4

Install this agent skill to your Project

npx add-skill https://github.com/Jamie-BitFlight/claude_skills/tree/main/plugins/dasel/skills/data-transformation

SKILL.md

Data Transformation with Dasel v3

<when_to_use>

Activate this skill when:

  • Modifying values in config files (YAML, TOML, JSON, XML, HCL, INI)
  • Converting between data formats (JSON to YAML, TOML to JSON, etc.)
  • Batch-updating array elements (prices, versions, flags)
  • Appending, removing, or reshaping data structures
  • Merging objects or constructing new structures from existing data

</when_to_use>

Safety Rules

NEVER redirect output to the same input file. This truncates the file before dasel reads it, resulting in data loss:

bash
# WRONG — destroys input
dasel -f config.yaml --root 'port = 9090' > config.yaml

# CORRECT — write to temp, then rename
dasel -f config.yaml --root 'port = 9090' > config_tmp.yaml && mv config_tmp.yaml config.yaml

Always verify before overwriting. Preview the transformation output first, then redirect:

bash
# Preview
dasel -f config.yaml --root 'server.port = 9090'
# Apply
dasel -f config.yaml --root 'server.port = 9090' > config_tmp.yaml && mv config_tmp.yaml config.yaml

Core Modification Syntax

In dasel v3, put and delete subcommands do not exist. All modifications use assignment expressions with --root to output the full document.

Set a Value

bash
# Output full document with one field changed
dasel -f config.yaml --root 'server.port = 9090'

# Numeric, boolean, and string assignments
echo '{"count": 1}' | dasel -i json --root 'count = 42'
echo '{"enabled": false}' | dasel -i json --root 'enabled = true'
echo '{"name": "old"}' | dasel -i json --root 'name = "new"'

Set Nested Values

bash
dasel -f config.yaml --root 'database.connection.host = "db.example.com"'
dasel -f config.yaml --root 'database.connection.port = 5432'

In-Place Modification (Safe Pattern)

bash
# Single field update
dasel -f config.yaml --root 'server.port = 9090' > config_tmp.yaml && mv config_tmp.yaml config.yaml

# Multiple fields — chain with semicolons
dasel -f config.yaml --root 'server.port = 9090; server.host = "0.0.0.0"' > config_tmp.yaml && mv config_tmp.yaml config.yaml

Format Conversion

Pipe through dasel with different input/output format flags:

bash
# JSON to YAML
cat data.json | dasel -i json -o yaml > data.yaml

# YAML to TOML
cat config.yaml | dasel -i yaml -o toml > config.toml

# JSON to XML
cat data.json | dasel -i json -o xml > data.xml

# TOML to JSON
cat config.toml | dasel -i toml -o json > config.json

# CSV to JSON
cat data.csv | dasel -i csv -o json > data.json

Array Operations

Append to Array

bash
# Add element to end of array
echo '[1,2,3]' | dasel -i json --root '[$this..., 4]'
# Output: [1, 2, 3, 4]

# Append object to array
dasel -f data.json --root 'items = [$root.items..., {"name": "new", "value": 42}]'

Batch Update with each

bash
# Multiply all prices by 1.1 (10% increase)
dasel -f data.json --root 'items.each(price = price * 1.1)'

# Set a flag on all elements
dasel -f data.json --root 'users.each(active = true)'

# Increment all values
echo '[1,2,3]' | dasel -i json --root 'each($this = $this + 1)'

Filter Then Transform

bash
# Get only active users, then extract names
dasel -f data.json 'users.filter(active == true).map(name)'

Object Operations

Merge via Spread

bash
# Add new key to existing object
dasel -f base.json --root '{ $root..., "newKey": "value" }'

# Merge two objects
dasel -f base.json --root '{ $root..., "extra": true, "version": 2 }'

Field Deletion via Reconstruction

Since v3 has no delete command, remove fields by constructing a new object with only the desired keys:

bash
# Keep only "name" and "email", drop everything else
echo '{"name":"a","email":"b","password":"c"}' | dasel -i json --root '{ name, email }'

Field Rename via Reconstruction

bash
echo '{"old_name": "value"}' | dasel -i json --root '{ "new_name": old_name }'

Multi-Step Transformations

Use variable assignment and semicolons for complex operations:

bash
# Store intermediate result in variable, then transform
dasel -f data.json '$active = users.filter(active == true); $active.map(name)'

# Multiple variables
dasel -f data.json '$a = items.filter(price > 100); $b = $a.map(name); $b'

Conditional Transformation

bash
# Set value based on condition
dasel -f data.json --root 'if(count > 5) { status = "many" } else { status = "few" }'

Transformation Patterns

For detailed per-use-case transformation patterns (config updates, batch processing, format migration, data reshaping, error handling), see Transformation Patterns.

References

Didn't find tool you were looking for?

Be as detailed as possible for better results