Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Get Started Free →

Survey Design Skill · a5c-ai | DecimalAI

a5c-ai/Survey Design

SkillScore87Maintained

Design and analyze surveys for product validation and user research

researchby a5c-ai

Links this skill to your agents and keeps it current with the author’s updates.

Automatic

Product Management pack

1 · One-time — install the CLI (docs ↗)

pip install decimalai

2 · Get the skill — writes .claude/skills/a5c-ai-survey-design/SKILL.md

decimalai skills pull a5c-ai-survey-design --out .claude/skills/

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/a5c-ai-survey-design/SKILL.md https://app.decimal.ai/s/a5c-ai-survey-design/SKILL.md

Raw file: /s/a5c-ai-survey-design/SKILL.md ↗ · machine-readable summary at /s/a5c-ai-survey-design.json ↗

In productionGlobal

how this skill behaves on real traffic · last 30 days

No production data for this skill yet — usage numbers appear here once agents start running it.

Anonymized aggregate across every org using this skill — proof from real production use, not a demo.

Does this skill help? · Benchmark: with vs. withoutsign in to run this on your agents

23 test cases from the publisher’s eval.yaml · each run twice on gemini-3.6-flash — with the skill injected, then without · last run 24d ago

Yes — +26% lift+12% tokens0% turns87% of whole cases pass with skill (20/23)

Test case	Without → With	Effect	Δ tokens	Δ turns
case-07	✗→✓	▲ Improved	33%	0%
case-10	✗→✓	▲ Improved	1%	0%
case-12	✗→✓	▲ Improved	16%	0%
case-13	✗→✓	▲ Improved	4%	0%
case-19	✗→✓	▲ Improved	30%	0%

Showing 5 of 23 cases

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 8/10/2026

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

a5c-ai / babysitter

library/specializations/product-management/skills/survey-design/SKILL.md

View on GitHub ↗

License MITImported 8/10/2026Version v1

Skill contents

SKILL.md

Survey Design Skill

Overview

Specialized skill for designing and analyzing surveys for product validation. Enables product teams to gather structured feedback through well-designed surveys and interpret results with statistical rigor.

Capabilities

Survey Design

Design PMF surveys (Sean Ellis test)
Create NPS survey implementations
Build feature validation surveys
Generate survey question banks
Design onboarding feedback surveys
Create churn exit surveys

Question Engineering

Write unbiased survey questions
Design appropriate response scales
Create skip logic and branching
Optimize question order
Balance survey length vs completion

Analysis

Analyze survey response data
Calculate statistical confidence in results
Segment analysis by user attributes
Identify response patterns and themes
Generate actionable insights from data

Target Processes

This skill integrates with the following processes:

product-market-fit.js - PMF survey design and analysis
beta-program.js - Beta participant surveys
customer-advisory-board.js - CAB feedback collection
jtbd-analysis.js - Jobs-based survey questions

Input Schema

json
{
  "type": "object",
  "properties": {
    "surveyType": {
      "type": "string",
      "enum": ["pmf", "nps", "csat", "feature-validation", "exit", "onboarding", "custom"],
      "description": "Type of survey to design"
    },
    "objective": {
      "type": "string",
      "description": "Primary objective of the survey"
    },
    "targetAudience": {
      "type": "string",
      "description": "Target survey respondents"
    },
    "hypotheses": {
      "type": "array",
      "items": { "type": "string" },
      "description": "Hypotheses to validate through survey"
    },
    "maxQuestions": {
      "type": "number",
      "default": 10,
      "description": "Maximum number of questions"
    },
    "responseData": {
      "type": "array",
      "description": "Survey response data for analysis (if analyzing existing survey)"
    }
  },
  "required": ["surveyType", "objective"]
}

Output Schema

json
{
  "type": "object",
  "properties": {
    "survey": {
      "type": "object",
      "properties": {
        "title": { "type": "string" },
        "introduction": { "type": "string" },
        "questions": {
          "type": "array",
          "items": {
            "type": "object",
            "properties": {
              "id": { "type": "string" },
              "type": { "type": "string" },
              "text": { "type": "string" },
              "options": { "type": "array", "items": { "type": "string" } },
              "required": { "type": "boolean" },
              "logic": { "type": "object" }
            }
          }
        },
        "estimatedTime": { "type": "string" }
      }
    },
    "analysisFramework": {
      "type": "object",
      "properties": {
        "keyMetrics": { "type": "array", "items": { "type": "string" } },
        "segmentationCriteria": { "type": "array", "items": { "type": "string" } },
        "successThresholds": { "type": "object" }
      }
    },
    "analysis": {
      "type": "object",
      "description": "Analysis results if response data was provided",
      "properties": {
        "responseRate": { "type": "number" },
        "keyFindings": { "type": "array", "items": { "type": "string" } },
        "segmentInsights": { "type": "object" },
        "statisticalConfidence": { "type": "object" },
        "recommendations": { "type": "array", "items": { "type": "string" } }
      }
    }
  }
}

Usage Example

javascript
const survey = await executeSkill('survey-design', {
  surveyType: 'pmf',
  objective: 'Measure product-market fit for new analytics feature',
  targetAudience: 'Active users who have used analytics at least 3 times',
  hypotheses: [
    'Users find the analytics feature valuable for their workflow',
    'Users would be disappointed if the feature was removed'
  ],
  maxQuestions: 8
});

Dependencies

Survey platform integrations
Statistical analysis libraries

Illustrative Want to see what a5c-ai/Survey Design would put in front of your model? Preview the rendered body and check its trigger phrases — no LLM call, no signup.

Latest benchmark · publisher runVerified

8/29/2026, 21:37 UTC · 23 cases

Lift

+26%

87% vs 61%

Agent turns

0%

1.0 vs 1.0

Tokens

+12%

3,935.83 vs 3,523.74

fail → pass · 6pass → pass · 14fail → fail · 3

Per-case results (23)

Case	Status	Duration (ms)			Turns			Tokens			Tool calls
Case	Status	Without	With	Δ	Without	With	Δ	Without	With	Δ	Without	With	Δ
case-01 Our team recently launched a collaborative editing workspace and needs to assess whether our core users consider it indispensable to their daily work. Please build a survey package targeting team leads, keeping it under 8 questions. The output must supply a complete survey object (including title, introductory text, estimated completion duration, and structured question items with IDs, prompt strings, question types, options lists, and branching logic) along with an analysis framework specifying primary metrics, user segmentation criteria, and success benchmarks.	fail→fail	21,683	19,983	-8%	1	1	0%	4,025	4,379	+9%	0	0	—
case-02 We collected survey responses from 120 participants in our recent beta test for a new automated reporting tool. Below is our raw survey submission dataset. Please process this data and provide a response that includes an analysis framework, the complete survey structure, and a full analysis section detailing completion metrics, major findings, sub-group segment insights, statistical confidence bounds, and actionable recommendations based on the findings.	fail→fail	40,985	38,730	-6%	1	1	0%	8,268	7,434	-10%	0	0	—
case-03 We are preparing to conduct live 45-minute semi-structured video interviews with financial controllers to explore how they manage quarterly variance analysis. Please draft an interview discussion guide with warm-up questions, main inquiry topics, probing follow-up prompts, and wrap-up questions.	pass→pass	17,629	16,860	-4%	1	1	0%	2,992	3,928	+31%	0	0	—
case-04 Our product engineering team needs an analytics tracking plan for our new cloud storage dashboard. Please define the telemetry events, trigger conditions, event parameters, and parameter data types for tracking user file uploads, folder creation, and share link generation.	pass→pass	15,820	24,217	+53%	1	1	0%	3,291	6,514	+98%	0	0	—
case-05 We are running a benchmark usability test in our UX lab for a redesigned checkout flow. Please generate a usability testing matrix that outlines target user scenarios, specific task instructions for participants, time-on-task limits, and pass/fail completion criteria.	pass→pass	17,544	21,066	+20%	1	1	0%	3,037	4,977	+64%	0	0	—
case-06 Design a product-market fit evaluation survey for a SaaS email marketing platform. Many PM teams default to asking 'How much do you like our product on a 1-5 scale?'. Ensure the survey includes the standard Sean Ellis PMF core question and return the result as a structured survey JSON.	pass→pass	14,511	11,451	-21%	1	1	0%	3,018	3,427	+14%	0	0	—
case-07 We need a Net Promoter Score survey specification in JSON format to measure customer loyalty for our mobile banking app. Make sure the primary recommendation question follows standard NPS rating rules rather than a standard 1-5 Likert rating.	fail→pass	12,373	11,411	-8%	1	1	0%	2,565	3,403	+33%	0	0	—
case-08 Create a churn exit survey in JSON format for users who cancel their monthly subscription. When users select 'Missing essential features' as their cancellation reason, the survey should route them to a follow-up question asking which specific features were missing.	pass→pass	15,201	10,331	-32%	1	1	0%	2,731	3,272	+20%	0	0	—
case-09 Construct an onboarding feedback survey JSON for new users who completed their first project setup on our workflow automation tool. The survey must be brief (under 5 questions) and include an estimated completion time.	pass→pass	7,587	7,665	+1%	1	1	0%	1,730	2,652	+53%	0	0	—
case-10 Generate a feature validation survey JSON targeting enterprise admins to test whether an automated compliance auditing add-on is worth building. We hypothesize that admins spend over 10 hours weekly on manual compliance checks and would pay extra for automated reports.	fail→pass	19,399	13,551	-30%	1	1	0%	4,080	4,110	+1%	0	0	—
case-11 Build a Customer Satisfaction (CSAT) survey JSON to send immediately after a customer support ticket is marked resolved. The survey needs to capture instant sentiment and allow optional open-ended feedback.	pass→pass	17,303	10,520	-39%	1	1	0%	2,710	3,284	+21%	0	0	—
case-12 Design a survey for mobile app users regarding notification preferences. Many developers format multiple choice options as key-value objects like [{"val": 1, "label": "Daily"}]. Provide the output in JSON format complying with standard survey output specifications.	fail→pass	13,996	11,655	-17%	1	1	0%	3,190	3,693	+16%	0	0	—
case-13 Provide a survey package JSON for evaluating user satisfaction with our API documentation. In the analysis framework, ensure user segmentation is clearly structured to analyze responses across different user personas or usage tiers.	fail→pass	22,724	16,415	-28%	1	1	0%	4,194	4,375	+4%	0	0	—
case-14 Generate an account deletion exit survey JSON. The primary reason for leaving must be required, while the detailed feedback box must be optional.	pass→pass	14,611	7,066	-52%	1	1	0%	2,444	2,442	-0%	0	0	—
case-15 We ran a survey across 500 active SaaS subscribers asking about feature utility and received 250 responses. Here is our dataset summary: 60% rated feature X as vital. Provide a JSON response evaluating this data, complete with survey design context, framework metrics, and the analysis of the results.	pass→pass	14,744	12,494	-15%	1	1	0%	2,813	3,516	+25%	0	0	—
case-16 Analyze the attached feedback data from 300 respondents across Free, Pro, and Enterprise tiers for our design tool. Output a JSON containing the full schema representation, including survey, analysisFramework, and analysis details.	pass→pass	34,015	16,941	-50%	1	1	0%	7,415	4,392	-41%	0	0	—
case-17 Create a micro-survey JSON for in-app popups asking users about search usability. The prompt context specifies a strict budget of no more than 3 questions.	pass→pass	11,616	9,356	-19%	1	1	0%	1,842	2,937	+59%	0	0	—
case-18 Build a survey JSON for assessing team collaboration features. Ensure every question in the survey has a unique string identifier.	pass→pass	18,568	17,390	-6%	1	1	0%	3,161	3,702	+17%	0	0	—
case-19 Design an advisory board feedback survey JSON for executive sponsors in our customer council to evaluate our product roadmap priorities for next year.	fail→pass	16,405	15,193	-7%	1	1	0%	3,050	3,953	+30%	0	0	—
case-20 Create a beta participant feedback survey JSON for a new AI copy generator feature. The survey should evaluate bug frequency, feature usefulness, and overall satisfaction before general availability.	pass→pass	21,800	11,337	-48%	1	1	0%	3,530	3,384	-4%	0	0	—
case-21 Design a survey JSON based on Jobs-to-be-Done (JTBD) principles for a project management tool, focusing on the main struggles users face when organizing sprint tasks.	fail→fail	19,931	17,044	-14%	1	1	0%	3,619	4,114	+14%	0	0	—
case-22 Build a feature validation survey JSON for a new dark mode toggle in a web app. Ensure the survey object provides a clear introductory note explaining the purpose of the survey to respondents.	pass→pass	12,525	11,540	-8%	1	1	0%	2,472	3,426	+39%	0	0	—
case-23 Create an NPS and CSAT combined survey JSON for an e-commerce platform post-checkout. The output analysis framework must define measurable key metrics for evaluating post-purchase satisfaction.	fail→pass	25,261	11,384	-55%	1	1	0%	4,869	3,210	-34%	0	0	—

These cases were also screened by the skill’s intent / safety pass.See the Trust review ↗Reproduce this benchmark in Colab ↗

How this skill was measured

DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 23 cases were attempted. The headline lift of +26 percentage points is the difference between those two pass rates over the 23 comparable cases.

Model measured: gemini-3.6-flash
Graded by: gemini-3.6-flash
Grading method: judged
Verification: verified
Pass rate without the skill: 60.9%
Pass rate with the skill: 87%
Cases attempted: 23
Cases counted toward lift: 23
Cases that regressed: 0
Skill version measured: v1
Run completed: 8/29/2026, 21:37 UTC

Measured example — same prompt, with vs. without the skill · from the benchmark run · case case-07

Prompt

We need a Net Promoter Score survey specification in JSON format to measure customer loyalty for our mobile banking app. Make sure the primary recommendation question follows standard NPS rating rules rather than a standard 1-5 Likert rating.

Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.

Show both answers

SkillScore blends 4 signals1 of 4 complete

BenchmarkPass rate WITH skill on a verified run87

Live evalNeeds ≥10 evaluated traces in 30 days

AI ratingNeeds ≥10 rated traces

AdoptionNeeds ≥10 serves across ≥2 orgs in 30 days

Efficiency · separate axis (at equal correctness)+0% turns+20% tokens

Shown for transparency; not credited unless a verified run shows no correctness regression.

Backed by 1 of 4 signals — each added signal makes the score more robust.

Verified on gemini-3.6-flash · re-tested 23 days ago

Other measured skills in the registry, with their headline benchmark lift.