| name | output-eval-dataset-design |
| description | Design diverse eval datasets using dimension-based variation. Use when bootstrapping eval datasets, when real traces are sparse, or when existing datasets miss edge cases. |
| allowed-tools | ["Bash","Read","Write","Edit"] |
Designing Eval Datasets
Overview
Diverse datasets catch more failures. This skill teaches dimension-based dataset design — systematically varying inputs along axes that target failure-prone regions of your workflow. The output is a set of YAML dataset files ready for output workflow test.
When to Use
- Bootstrapping an eval dataset for a new workflow
- Existing datasets only cover happy paths
- Real production traces are sparse (fewer than 50)
- Stress-testing specific failure hypotheses from error analysis
When NOT to Use
- You already have 100+ representative real traces — use stratified sampling from those instead of generating synthetic data
- You haven't done error analysis yet — do that first (
output-eval-error-analysis) so your dimensions target real failure modes, not guesses
Step 1: Define Dimensions
Identify 3+ axes of input variation that target anticipated failure modes. Each dimension should vary one aspect of the input that you expect to influence output quality.
Deriving dimensions from error analysis
Map failure categories to input properties that trigger them:
| Failure Category | Triggering Input Property | Dimension |
|---|
| Off-topic drift | Ambiguous or broad topics | Topic specificity: specific / broad / ambiguous |
| Tone mismatch | Conflicting tone signals | Tone difficulty: simple / nuanced / contradictory |
| Too short | Short or vague input | Input detail: minimal / moderate / comprehensive |
| Missing sections | Many explicit requirements | Requirement count: 0 / 1-2 / 5+ |
| Hallucinated URLs | Technical topics with real entities | Entity density: none / few / many |
Example dimensions for a blog generation workflow
| Dimension | Values | Why |
|---|
| Topic complexity | simple, technical, ambiguous | Technical and ambiguous topics trigger more hallucination and drift |
| Tone request | none, formal, casual, contradictory | Explicit tone requests reveal tone-matching failures |
| Length constraint | none, short (100w), long (1000w) | Extreme length constraints trigger truncation and padding |
| Required sections | none, 1 section, 3+ sections | Multiple required sections stress structural compliance |
Aim for 3-5 dimensions. More than 5 creates an unmanageable combinatorial space.
Step 2: Draft Tuples
Create ~20 combinations of dimension values. Cover the extremes and the combinations most likely to cause failures.
Tuple selection strategy
- Cover every dimension value at least twice — ensures no blind spots
- Pair difficult values together — "ambiguous topic + contradictory tone + 3+ required sections" is where failures cluster
- Include a few easy combinations — confirms the workflow works under normal conditions
- Avoid near-duplicates — each tuple should test a distinct scenario
Example tuples
| # | Topic Complexity | Tone | Length | Required Sections |
|---|
| 1 | simple | none | none | none |
| 2 | simple | formal | short | 1 section |
| 3 | technical | formal | long | 3+ sections |
| 4 | technical | casual | short | none |
| 5 | ambiguous | formal | long | 1 section |
| 6 | ambiguous | contradictory | none | 3+ sections |
| 7 | simple | contradictory | long | none |
| 8 | technical | none | none | 3+ sections |
| 9 | ambiguous | casual | short | 1 section |
| 10 | simple | casual | long | 3+ sections |
| 11 | technical | formal | short | 1 section |
| 12 | ambiguous | none | long | none |
| 13 | simple | formal | none | 3+ sections |
| 14 | technical | contradictory | long | 1 section |
| 15 | ambiguous | formal | short | 3+ sections |
| 16 | simple | none | short | 1 section |
| 17 | technical | casual | long | 3+ sections |
| 18 | ambiguous | contradictory | short | none |
| 19 | technical | formal | none | none |
| 20 | ambiguous | casual | long | 3+ sections |
Review each tuple and ask: "Is this a realistic scenario a user might create?" Discard any that aren't.
Step 3: Convert Tuples to Workflow Inputs
Transform each tuple into a JSON object matching the workflow's inputSchema from types.ts.
Read the schema first
cat src/workflows/<workflow_name>/types.ts
Manual conversion (small datasets)
For each tuple, write the corresponding JSON input:
Tuple 3: technical + formal + long + 3+ sections
{
"topic": "Quantum error correction techniques in superconducting qubit architectures",
"tone": "formal",
"min_length": 1000,
"required_sections": ["Introduction", "Technical Background", "Current Approaches", "Challenges", "Conclusion"]
}
Tuple 6: ambiguous + contradictory + none + 3+ sections
{
"topic": "Things that matter",
"tone": "Write in a formal academic style but keep it super casual and fun",
"required_sections": ["Overview", "Deep Dive", "Takeaways"]
}
Use realistic, natural-sounding inputs. Avoid test-looking data like "Test topic 1" or "Lorem ipsum."
LLM-assisted conversion (larger datasets)
For 20+ tuples, use an LLM to batch-convert tuples into realistic inputs. Create a prompt that takes the tuple values and the inputSchema, then generates a natural JSON input. Review every generated input for realism before using it.
Step 4: Generate Dataset Files
Run each input through the workflow to capture real output:
npx output workflow dataset generate blog_generator \
--input '{"topic": "Quantum error correction", "tone": "formal", "min_length": 1000}' \
--name technical_formal_long
npx output workflow dataset generate blog_generator \
--input '{"topic": "Things that matter", "tone": "Write formally but keep it casual"}' \
--name ambiguous_contradictory
Each command creates a YAML file in tests/datasets/ with input and last_output populated.
Naming convention
Use names that encode the key dimensions:
tests/datasets/
├── simple_no_constraints.yml
├── simple_formal_short.yml
├── technical_formal_long_sections.yml
├── technical_casual_short.yml
├── ambiguous_formal_long.yml
├── ambiguous_contradictory_sections.yml
└── ...
Batch generation
For many datasets, create a shell script:
#!/bin/bash
inputs=(
'{"topic": "Solar panels", "tone": "formal"}|simple_formal'
'{"topic": "Quantum error correction in superconducting qubits", "tone": "casual", "min_length": 1000}|technical_casual_long'
'{"topic": "Things that matter", "required_sections": ["Overview", "Dive", "Takeaways"]}|ambiguous_sections'
)
for entry in "${inputs[@]}"; do
IFS='|' read -r input name <<< "$entry"
echo "Generating: $name"
npx output workflow dataset generate blog_generator --input "$input" --name "$name"
done
Step 5: Add Ground Truth
After generation, edit each dataset YAML to add ground_truth labels. Review the last_output and assign verdicts per evaluator.
name: ambiguous_contradictory_sections
input:
topic: "Things that matter"
tone: "Write in a formal academic style but keep it super casual and fun"
required_sections: ["Overview", "Deep Dive", "Takeaways"]
last_output:
output:
title: "Things That Matter: A Casual Academic Exploration"
blog_post: "So here's the thing about stuff that matters..."
executionTimeMs: 4200
date: '2026-03-25T00:00:00.000Z'
ground_truth:
human_verdict: fail
notes: "Contradictory tone caused drift; missing 'Deep Dive' section"
evals:
check_tone:
expected_tone: "formal academic style"
verdict: fail
check_topic:
required_topic: "Things that matter"
verdict: partial
check_sections:
required_sections: ["Overview", "Deep Dive", "Takeaways"]
verdict: fail
check_length:
min_length: 100
verdict: pass
Labeling tips
- Label the global
human_verdict for every dataset (pass/fail)
- Label the top 3 evaluators by failure rate first
- For borderline cases, decide and commit — don't leave ambiguous labels
- Record
notes explaining your reasoning for future reference
Step 6: Supplement with Real Data
If you have production traces available, use them to fill coverage gaps.
npx output workflow dataset generate blog_generator --download --limit 30
Stratified sampling
After downloading, check which dimensions are underrepresented:
- Review the downloaded datasets
- Tag each with its approximate dimension values
- Identify gaps (e.g., no ambiguous topics, no contradictory tones)
- Generate synthetic datasets specifically for the missing combinations
Mixing real and synthetic
A good eval dataset combines both:
- Real traces (60-70%) — capture authentic user behavior and edge cases you didn't anticipate
- Synthetic traces (30-40%) — fill coverage gaps and stress-test specific failure regions
Step 7: Quality Filter
Review every dataset before using it in evals.
Discard datasets where
- The input is unrealistic or test-looking
- The input is nearly identical to another dataset (redundant)
- The workflow errored out entirely (test infrastructure issues, not quality issues)
- The ground truth labels are ambiguous or contested
Target dataset count
| Workflow Complexity | Minimum Datasets | Target |
|---|
| Simple (1-2 steps, narrow input) | 10 | 20 |
| Medium (3-5 steps, moderate variation) | 20 | 40 |
| Complex (5+ steps, wide input space) | 40 | 80+ |
More datasets improve eval reliability. Aim for at least 5 datasets per failure category to get meaningful failure rates.
Verification
npx output workflow dataset list blog_generator
npx output workflow test blog_generator --cached
npx output workflow test blog_generator --cached --dataset technical_formal_long,ambiguous_contradictory
Check that:
Related Skills
output-eval-error-analysis — Identify failure modes that inform dimension selection
output-dev-eval-testing — Dataset YAML format, CLI commands, evalWorkflow() setup
output-dev-scenario-file — Scenario JSON files as seed inputs for dataset generation
output-eval-validate-judge — Split datasets into train/dev/test for judge validation
output-eval-judge-prompt — Design LLM judge prompts that consume these datasets