test-idea-rewriting — Claude Code skill: what it does, how to install
test-idea-rewriting is a Claude Code skill that turns passive test descriptions into active actions with observable results. It runs when test ideas are vague or fail quality validation.
MIT 104 installs ★ 533 updated 2026-10-25
npx skills add proffesor-for-testing/agentic-qe --skill test-idea-rewriting -a claude-codeWhat test-idea-rewriting does
The skill replaces vague test phrases starting with verify into action-verb patterns with observable results. It scans inputs using regular expressions, maps statements to interaction or measurement categories, and preserves existing identifiers and priorities.
Operates across API, UI, database, and performance test suites. Executes single-file and batch transformations through the qe-test-idea-rewriter agent while maintaining formatting and metadata.
What it helps with
- transform passive test phrasing into action verbs
- remove vague verify patterns from test suites
- preserve metadata during test idea refactoring
- convert API checks into observable requests with metrics
Example prompts for the agent
- Rewrite test ideas in
assessment.htmlusing the test-idea-rewriting skill - Batch process test files in
./assessments/directory - Run quality checks on test cases with
qe-test-idea-rewriter
How to install test-idea-rewriting
1Skills CLI (Vercel)
npx skills add proffesor-for-testing/agentic-qe --skill test-idea-rewriting -a claude-code2Manually: put the files in place
~/.claude/skills/test-idea-rewriting/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/test-idea-rewriting/
1Skills CLI (Vercel)
npx skills add proffesor-for-testing/agentic-qe --skill test-idea-rewriting -a codex2Manually: put the files in place
.agents/skills/test-idea-rewriting/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add proffesor-for-testing/agentic-qe --skill test-idea-rewriting -a cursor2Manually: put the files in place
.cursor/skills/test-idea-rewriting/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install proffesor-for-testing/agentic-qe assets/skills/test-idea-rewriting2Manually: put the files in place
.github/skills/test-idea-rewriting/SKILL.md1Skills CLI (Vercel)
npx skills add proffesor-for-testing/agentic-qe --skill test-idea-rewriting -a gemini-cli2Platform CLI
gemini skills install https://github.com/proffesor-for-testing/agentic-qe --path assets/skills/test-idea-rewritingManually: put the files in place
.devin/skills/test-idea-rewriting/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add proffesor-for-testing/agentic-qe --skill test-idea-rewriting -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about test-idea-rewriting
How to install test-idea-rewriting in Claude Code?
Run npx skills add proffesor-for-testing/agentic-qe --skill test-idea-rewriting -a claude-code and check ~/.claude/skills/test-idea-rewriting/SKILL.md.
What does test-idea-rewriting do?
It detects passive verify patterns in test ideas and rewrites them into explicit action-outcome pairs for better testability.
Which agent handles the transformation?
The skill delegates tasks to the qe-test-idea-rewriter agent for file and batch transformations.
Source text
SKILL.md · MIT
--- name: test-idea-rewriting description: "Transform passive 'Verify X' test descriptions into active, observable test actions. Use when test ideas lack specificity, use vague language, or fail quality validation. Converts to action-verb format for clearer, more testable descriptions." category: test-design priority: medium tokenEstimate: 800 agents: [qe-test-idea-rewriter] implementation_status: optimized optimization_version: 1.0 last_optimized: 2026-01-17 dependencies: [] quick_reference_card: true tags: [test-ideas, rewriting, action-verbs, quality, transformation] --- # Test Idea Rewriting <default_to_action> When transforming test ideas: 1. DETECT all "Verify X" patterns via regex 2. IDENTIFY appropriate action verb category 3. TRANSFORM to "[ACTION] [trigger]; [OBSERVE] [result]" pattern 4. PRESERVE all metadata (IDs, priorities, automation types) 5. VALIDATE zero "Verify" patterns remain 6. OUTPUT in same format as input **Success Criteria:** `/<td>Verify\s/gi` returns 0 matches </default_to_action> ## Quick Reference Card ### Transformation Pattern ``` [ACTION VERB] [specific trigger]; [OUTCOME VERB] [observable result] ``` ### Action Verb Quick Reference | Category | Verbs | Use When | |----------|-------|----------| | **Interaction** | Click, Type, Submit, Navigate, Scroll | UI actions | | **Trigger** | Send, Inject, Force, Simulate, Load | API/system actions | | **Measurement** | Measure, Time, Count, Profile | Performance checks | | **State** | Set, Configure, Enable, Disable, Toggle | Setup actions | | **Observation** | Confirm, Assert, Check, Observe | Outcome verification | ### Common Transformations | Before | After | |--------|-------| | Verify login works | Submit valid credentials; confirm session created | | Verify API returns 200 | Send GET request; assert 200 response within 500ms | | Verify error displays | Trigger validation error; observe error message | | Verify data saves | Insert record; query database; confirm fields match | | Verify performance | Execute 100 requests; measure p99 < 200ms | --- ## Transformation Rules ### Pattern Detection ```regex /<td>Verify\s/gi // HTML table cells /^Verify\s/gim // Line starts /"Verify\s[^"]+"/gi // Quoted strings ``` ### Transformation Categories #### API/Network Tests | Input Pattern | Output Pattern | |---------------|----------------| | Verify API returns X | Send [METHOD] request; assert [STATUS] response | | Verify endpoint accepts Y | Post [PAYLOAD] to endpoint; confirm [RESPONSE] | | Verify webhook fires | Trigger [EVENT]; observe webhook received | #### UI/UX Tests | Input Pattern | Output Pattern | |---------------|----------------| | Verify button works | Click [BUTTON]; observe [EFFECT] | | Verify form submits | Fill [FIELDS]; submit form; confirm [RESULT] | | Verify navigation works | Click [LINK]; observe [PAGE] loads | #### Data Tests | Input Pattern | Output Pattern | |---------------|----------------| | Verify data saves | Insert [RECORD]; query; confirm [MATCH] | | Verify validation works | Enter [INVALID]; observe [ERROR] | | Verify format accepted | Submit [FORMAT]; confirm [PROCESSED] | #### Performance Tests | Input Pattern | Output Pattern | |---------------|----------------| | Verify performance is good | Execute [LOAD]; measure [METRIC] < [THRESHOLD] | | Verify scalability | Increase [USERS] to [N]; monitor [RESOURCE] | | Verify timeout works | Inject [DELAY]; confirm timeout after [TIME] | --- ## Action Verb Reference ### Interaction Verbs | Verb | When to Use | Example | |------|-------------|---------| | Click | UI element interaction | Click "Submit" button | | Type | Text entry | Type "[email protected]" | | Submit | Form completion | Submit registration form | | Navigate | Page changes | Navigate to /settings | | Scroll | Viewport movement | Scroll to page bottom | | Drag | Drag-and-drop | Drag file to upload zone | | Hover | Mouse positioning | Hover over tooltip trigger | | Select | Dropdown/checkbox | Select "Admin" from role dropdown | ### Trigger Verbs | Verb | When to Use | Example | |------|-------------|---------| | Send | HTTP requests | Send POST to /api/orders | | Inject | Fault injection | Inject 500ms latency | | Force | State manipulation | Force offline mode | | Simulate | Event generation | Simulate device rotation | | Load | Resource loading | Load 50MB test file | | Execute | Script/command | Execute database migration | | Invoke | Function/webhook | Invoke payment callback | | Trigger | Event firing | Trigger scheduled job | ### Measurement Verbs | Verb | When to Use | Example | |------|-------------|---------| | Measure | Quantitative check | Measure response time | | Time | Duration tracking | Time page render | | Count | Quantity check | Count search results | | Profile | Resource analysis | Profile CPU usage | | Benchmark | Comparison | Benchmark against v1.0 | | Capture | State recording | Capture network traffic | | Monitor | Ongoing observation | Monitor memory for 5 minutes | ### Observation Verbs | Verb | When to Use | Example | |------|-------------|---------| | Confirm | Boolean check | Confirm user is logged in | | Assert | Value comparison | Assert total equals $99.99 | | Check | State verification | Check cart has 3 items | | Observe | Behavior watching | Observe spinner appears | | Validate | Rule compliance | Validate email format | | Expect | Predicted outcome | Expect redirect to /home | | Verify (avoid) | Use alternatives | Use confirm/assert instead | --- ## Quality Validation ### Pre-Transform Checks 1. Count "Verify" patterns in input 2. Identify context for each pattern 3. Map to appropriate action verb category ### Post-Transform Checks 1. Regex validation: zero "Verify" matches 2. Every test idea starts with action verb 3. Each test includes observable outcome 4. All metadata preserved unchanged ### Validation Regex ```javascript // Must return 0 matches for success const verifyPattern = /<td>Verify\s/gi; const matches = content.match(verifyPattern); if (matches && matches.length > 0) { throw new Error(`${matches.length} "Verify" patterns remain`); } ``` --- ## Agent Integration ```typescript // Single file transformation await Task("Rewrite Test Ideas", { inputFile: "assessment.html", outputFile: "assessment-rewritten.html", preserveFormatting: true }, "qe-test-idea-rewriter"); // Batch transformation await Task("Batch Rewrite", { inputDir: "./assessments/", outputDir: "./assessments-clean/", pattern: "*.html" }, "qe-test-idea-rewriter"); ``` --- ## Memory Namespace ``` aqe/rewriting/ ├── transformations/* - Transformation logs ├── patterns/* - Learned patterns └── vocabulary/* - Custom verb mappings ``` --- ## Related Skills - [sfdipot-product-factors](../sfdipot-product-factors/) - Assessment generation - [test-design-techniques](../test-design-techniques/) - Proper test structuring - [brutal-honesty-review](../brutal-honesty-review/) - Quality validation --- ## Remember **Every test idea should be actionable.** "Verify X works" tells you nothing about HOW to test. "[Action] X; [Observe] Y" gives clear steps and expected outcomes. Transform passive descriptions into active, observable tests.
© proffesor-for-testing · MIT · proffesor-for-testing/agentic-qe