Skill Evaluations Authoring

The Evaluation Template DSL Manual

Define tryout rubrics the way coaches think about the game — categories, skills, and scored metrics in one YAML document. Edit in the DSL editor with live preview; Apply DSL syncs the full template tree to your organization.

statshub.app
Evaluation Template DSL Editor — YAML authoring with live preview showing categories, skills, and 0–10 metric scales

If you can write a scouting checklist, you can author an evaluation template. The DSL is the checklist — the editor just validates it as you type.

Stats Hub Evaluation Template DSL (v1)

DSL building blocks

Every evaluation template is a four-level tree: template metadata, categories (position groups), skills (drill areas), and metrics (what coaches score).

📂

Categories

Top-level groupings — Pitcher, Positional, Catcher, or any structure your program uses. Order in the YAML is the order evaluators see drills.

categories: - name: Pitcher skills: [...]

Position / role groups

EvaluationsTryout organizers
🎯

Skills

Drill or competency areas within a category — Hitting, Infield, Pitch Speed. Each skill holds the metrics coaches score during that station.

- name: Hitting metrics: [Contact, Power, Plate Discipline]

Skill block syntax

EvaluationsStation design
📊

Metrics & scales

What gets scored. Default is subjective 0–10. Use shorthand (just a name) or specify custom ranges and measured types for velocity, pop time, etc.

- Contact - { name: "Fastball Velocity", scale: "60-100", type: measured }

Shorthand vs custom scale

EvaluationsScoring rubrics
👁

Live preview

The editor parses YAML on every keystroke and renders a tree with category/skill/metric counts, scale badges, and inline diagnostics.

2 categories · 7 skills · 14 metrics

Real-time validation

EvaluationsSplit-view authoring

Apply DSL

One click reconciles the YAML onto your database template — upserts by name, preserves IDs for metrics already used in events, guards deletions when scores exist.

POST /templates/[id]/apply-dsl

Name-based reconciliation

EvaluationsSafe publishing
🤖

Agent-friendly

YAML is the lingua franca for AI assistants. Paste a generated rubric, export .eval.yaml for version control, or let an agent draft categories from a scouting sheet.

template: Baseball Tryout sport: baseball

Human + machine readable

EvaluationsAutomation

From rubric to live tryout

The DSL editor in the screenshot above replaces dozens of form fields. One document defines everything coaches score on gameday.

1

Create or open a template

Go to Admin → Skill Evaluations → Templates. Create a preset (seeds Pitcher / Positional / Catcher) or open an existing template and click Edit DSL.

2

Author categories and skills

Structure your tryout stations as YAML. Each category is a position group; each skill is a drill area. Add skills under the right category — order matters for evaluator navigation.

3

Define metrics and scales

List metrics as simple names for standard 0–10 subjective scores. For measured values (velocity, 60 time), use scale: "min-max" and type: measured.

4

Validate and apply

Watch the live preview and fix any warnings (e.g. empty metric lists). Click Apply DSL to sync. If metrics with existing scores would be removed, the editor prompts for force-apply.

5

Attach to an evaluation event

Create an event, pick your template, assign players. Evaluators open the portal on their phones and score every metric you defined — no template changes mid-event.

Syntax reference

Tab through quick-start, grammar, metric scales, and a full baseball tryout example. Comments start with # — blank lines are ignored.

# Evaluation Template DSL — 60-second start template: Baseball Tryout sport: baseball categories: - name: Pitcher skills: - name: Mechanics metrics: [Delivery, Arm Slot, Balance] - name: Velocity metrics: - Fastball Velocity - Command

Paste into the DSL Editor — preview updates instantly. Click Apply DSL to save.

Ready to run tryouts with a rubric everyone understands?

Author templates in YAML, apply once, and let coaches score from their phones — same metrics, same scales, every evaluator.