Data Keywords

Three keywords for structured data — tables, rows, and quantitative metrics.

headers:

Category: Data Arabic: أعمدة:

Table header row — declares the column labels for the following row: blocks.

Syntax

headers: Col1 | Col2 | Col3

Examples

headers: Department | Budget | Spent | Remaining
row: Engineering | USD 500,000 | USD 320,000 | USD 180,000
row: Marketing | USD 200,000 | USD 175,000 | USD 25,000
row: Operations | USD 150,000 | USD 98,000 | USD 52,000

Notes

• headers: declares the column labels; row: provides the data
• Write headers: — columns: is not reserved; it parses as your own custom block
• Tables can also be written using Markdown pipe syntax (see below)

---

row:

Category: Data Arabic: صف:

Table data row — pipe-separated cell values. Must follow a headers: declaration.

Syntax

row: Cell1 | Cell2 | Cell3

Examples

headers: Name | Role | Email
row: Ahmed Al-Rashid | CEO | ahmed@acme.com
row: Sarah Chen | General Counsel | sarah@acme.com

---

Markdown pipe tables

Tables can also be written using standard Markdown pipe syntax. The parser produces the same internal table structure.

| Header 1 | Header 2 | Header 3 |
| Cell 1   | Cell 2   | Cell 3   |
| Cell 4   | Cell 5   | Cell 6   |

Column alignment

A separator row under the headers aligns its columns, as in GitHub Markdown: :--- left, :---: center, ---: right, --- the default (the document's start side). It works under a headers: line too:

headers: # | Description | Qty | Total
|---|---|---:|---:|
| 1 | Office chair | 12 | 3,000.00 |
| 2 | Meeting table | 3 | 11,245.00 |

The alignment is physical, as in GFM, and the header cell follows its column. A right-aligned column also uses tabular figures, so money lines up on its decimal point in Arabic and English alike. The separator row is kept on serialization (one with no colon carries nothing and is dropped, as before).

Dynamic table (templates)

The each: property on the header row enables dynamic row expansion:

| Description          | Qty          | Rate          | Total          | each: items |
| {{item.description}} | {{item.qty}} | {{item.rate}} | {{item.total}} |

1. Parser reads the each: array from data
2. Auto-singularizes: items → item, orders → order
3. Explicit naming: each: orders as order
4. Expands one row per array element
5. Zero items = zero rows; header always present

---

metric:

Category: Data Arabic: مؤشر:

Quantitative measurement or KPI. Queryable, renderable as cards or grids, with optional target comparison and trend indicators.

Syntax

metric: label | value: number | unit: string | period: timeframe | as-of: date | target: number | trend: direction | owner: team | source: origin

Properties

Property	Type	Required	Description
value	string	yes	The measured value
unit	string	no	Unit of measurement (USD, %, hours, etc.)
period	string	no	Reporting period (Q1 2026, March 2026, etc.)
as-of	string	no	Date the measurement was taken
target	string	no	Target or goal value
trend	string	no	up, down, flat
owner	string	no	Responsible team or person
source	string	no	Data source

Examples

metric: Revenue | value: 1200000 | unit: USD | period: Q1 2026 | target: 1000000 | trend: up
metric: Customer Satisfaction | value: 4.6 | unit: /5 | period: March 2026 | target: 4.5
metric: Sprint Velocity | value: 42 | unit: points | period: Sprint 14 | target: 40 | trend: flat
metric: Subtotal | value: 24000 | unit: USD

---

Extension keywords

Typed I/O parameter blocks live in the stable x-form: namespace; assert / secret remain experimental (x-exp:).

Extension	Purpose
x-form: input	Declare a typed input / form-field parameter
x-form: output	Declare a typed output parameter
x-exp: assert	Testable assertion — evaluable by CI
x-exp: secret	Secret or credential reference (always redacted)

See the extensions overview in Keywords → for full syntax.