Managing Cucumber

Cucumber BDD Testing Management Skill

Manage and analyze Cucumber feature files, step definitions, and test results.

Core Helper Functions

#!/bin/bash

# Parse Cucumber JSON report
cucumber_results() {
    local report="${1:-cucumber-report.json}"
    cat "$report" 2>/dev/null | jq '{
        features: length,
        scenarios: [.[].elements[] | select(.type == "scenario")] | length,
        passed: [.[].elements[] | select(.type == "scenario") | select(all(.steps[].result.status; . == "passed"))] | length,
        failed: [.[].elements[] | select(.type == "scenario") | select(any(.steps[].result.status; . == "failed"))] | length
    }'
}

MANDATORY: Discovery-First Pattern

Always discover feature files and step definitions before querying specifics.

Phase 1: Discovery

#!/bin/bash

echo "=== Feature Files ==="
find . -name "*.feature" 2>/dev/null | grep -v node_modules | wc -l | xargs echo "Total feature files:"
find . -name "*.feature" 2>/dev/null | grep -v node_modules | head -15

echo ""
echo "=== Scenario Summary ==="
find . -name "*.feature" 2>/dev/null | grep -v node_modules | while read f; do
    SCENARIOS=$(grep -c "Scenario:" "$f" 2>/dev/null || echo 0)
    echo "$(basename $f)\t${SCENARIOS} scenarios"
done | column -t | head -15

echo ""
echo "=== Step Definition Files ==="
find . -name "*steps*" -o -name "*step_definitions*" 2>/dev/null | grep -v node_modules | head -15

echo ""
echo "=== Tags in Use ==="
grep -rh "@" --include="*.feature" 2>/dev/null | grep -oP "@\w+" | sort | uniq -c | sort -rn | head -15

Phase 2: Analysis

#!/bin/bash

echo "=== Test Results ==="
REPORT=$(find . -name "cucumber-report.json" -o -name "cucumber_report.json" 2>/dev/null | head -1)
if [ -n "$REPORT" ]; then
    cucumber_results "$REPORT"

    echo ""
    echo "=== Failed Scenarios ==="
    cat "$REPORT" | jq -r '
        .[].elements[] | select(.type == "scenario") |
        select(any(.steps[].result.status; . == "failed")) |
        "\(.name)\t\(.tags | map(.name) | join(","))"
    ' | column -t | head -15

    echo ""
    echo "=== Failed Steps ==="
    cat "$REPORT" | jq -r '
        .[].elements[].steps[] | select(.result.status == "failed") |
        "\(.keyword)\(.name)\t\(.result.error_message | split("\n")[0] // "unknown")"
    ' | head -10
fi

echo ""
echo "=== Undefined Steps ==="
if [ -n "$REPORT" ]; then
    cat "$REPORT" | jq -r '
        .[].elements[].steps[] | select(.result.status == "undefined") |
        "\(.keyword)\(.name)"
    ' | sort -u | head -10
fi

Output Rules

• TOKEN EFFICIENCY: Target 50 lines per output
• Parse JSON or HTML reports for structured data
• Never dump full feature files — summarize scenario names and tags

Anti-Hallucination Rules

• NEVER guess feature file names — always discover via filesystem
• NEVER fabricate scenario results — parse actual Cucumber reports
• NEVER assume step definitions exist — check for undefined steps

Safety Rules

• NEVER modify feature files or step definitions without user confirmation
• NEVER delete test reports without user approval
• NEVER change tag configurations without user consent

Output Format

Present results as a structured report:

Managing Cucumber Report
════════════════════════
Resources discovered: [count]

Resource       Status    Key Metric    Issues
──────────────────────────────────────────────
[name]         [ok/warn] [value]       [findings]

Summary: [total] resources | [ok] healthy | [warn] warnings | [crit] critical
Action Items: [list of prioritized findings]

Target ≤50 lines of output. Use tables for multi-resource comparisons.

Counter-Rationalizations