Ready to supercharge your AI agents?

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

Get Started Free →

Customer Feedback Aggregation Skill · a5c-ai | DecimalAI

a5c-ai/Customer Feedback Aggregation

SkillScore79Maintained

Aggregate and analyze customer feedback from multiple sources for product insights

sales-supportby 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-customer-feedback-aggregation/SKILL.md

decimalai skills pull a5c-ai-customer-feedback-aggregation --out .claude/skills/

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

curl -fsSL --create-dirs -o .claude/skills/a5c-ai-customer-feedback-aggregation/SKILL.md https://app.decimal.ai/s/a5c-ai-customer-feedback-aggregation/SKILL.md

Raw file: /s/a5c-ai-customer-feedback-aggregation/SKILL.md ↗ · machine-readable summary at /s/a5c-ai-customer-feedback-aggregation.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

24 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 — +33% lift · at higher cost+37% tokens0% turns79% of whole cases pass with skill (19/24)

Test case	Without → With	Effect	Δ tokens	Δ turns
case-07	✗→✓	▲ Improved	236%	0%
case-10	✗→✓	▲ Improved	-21%	0%
case-12	✗→✓	▲ Improved	43%	0%
case-13	✗→✓	▲ Improved	66%	0%
case-16	✗→✓	▲ Improved	0%	0%

Showing 5 of 24 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/feedback-aggregation/SKILL.md

View on GitHub ↗

License MITImported 8/10/2026Version v1

Skill contents

SKILL.md

Customer Feedback Aggregation Skill

Overview

Specialized skill for aggregating and analyzing customer feedback from multiple sources. Enables product teams to synthesize voice-of-customer data into actionable insights for product decisions.

Capabilities

Data Collection

Parse support tickets for feature requests
Analyze NPS/CSAT verbatim responses
Extract themes from sales call notes
Monitor app store reviews
Aggregate feedback from Intercom/Zendesk
Process customer interview transcripts

Analysis

Calculate feature request frequency
Track sentiment trends over time
Identify emerging themes and patterns
Segment feedback by customer type
Correlate feedback with customer attributes
Detect urgency and impact signals

Synthesis

Generate feedback summary reports
Create feature request rankings
Build customer pain point matrices
Generate insight recommendations
Create feedback-to-feature mapping

Target Processes

This skill integrates with the following processes:

jtbd-analysis.js - Voice of customer for jobs analysis
feature-definition-prd.js - Customer-driven requirements
rice-prioritization.js - Reach and impact scoring
customer-advisory-board.js - CAB feedback synthesis

Input Schema

json
{
  "type": "object",
  "properties": {
    "sources": {
      "type": "array",
      "items": {
        "type": "object",
        "properties": {
          "type": { "type": "string", "enum": ["support-tickets", "nps-verbatim", "sales-calls", "app-reviews", "interviews", "surveys"] },
          "data": { "type": "array", "items": { "type": "object" } },
          "dateRange": { "type": "object" }
        }
      },
      "description": "Feedback data sources"
    },
    "analysisScope": {
      "type": "string",
      "enum": ["all", "feature-requests", "pain-points", "sentiment", "trends"],
      "description": "Focus area for analysis"
    },
    "segmentation": {
      "type": "array",
      "items": { "type": "string" },
      "description": "Dimensions to segment feedback by"
    },
    "timeRange": {
      "type": "object",
      "properties": {
        "start": { "type": "string", "format": "date" },
        "end": { "type": "string", "format": "date" }
      }
    }
  },
  "required": ["sources"]
}

Output Schema

json
{
  "type": "object",
  "properties": {
    "summary": {
      "type": "object",
      "properties": {
        "totalFeedbackItems": { "type": "number" },
        "sourceBreakdown": { "type": "object" },
        "dateRange": { "type": "object" },
        "overallSentiment": { "type": "string" }
      }
    },
    "themes": {
      "type": "array",
      "items": {
        "type": "object",
        "properties": {
          "theme": { "type": "string" },
          "frequency": { "type": "number" },
          "sentiment": { "type": "string" },
          "examples": { "type": "array", "items": { "type": "string" } },
          "segments": { "type": "object" }
        }
      }
    },
    "featureRequests": {
      "type": "array",
      "items": {
        "type": "object",
        "properties": {
          "feature": { "type": "string" },
          "requestCount": { "type": "number" },
          "customerSegments": { "type": "array", "items": { "type": "string" } },
          "urgencyScore": { "type": "number" },
          "impactEstimate": { "type": "string" },
          "representativeQuotes": { "type": "array", "items": { "type": "string" } }
        }
      }
    },
    "painPoints": {
      "type": "array",
      "items": {
        "type": "object",
        "properties": {
          "painPoint": { "type": "string" },
          "severity": { "type": "string" },
          "frequency": { "type": "number" },
          "customerImpact": { "type": "string" }
        }
      }
    },
    "trends": {
      "type": "object",
      "properties": {
        "emerging": { "type": "array", "items": { "type": "string" } },
        "declining": { "type": "array", "items": { "type": "string" } },
        "sentimentTrend": { "type": "string" }
      }
    },
    "recommendations": {
      "type": "array",
      "items": {
        "type": "object",
        "properties": {
          "recommendation": { "type": "string" },
          "priority": { "type": "string" },
          "evidence": { "type": "array", "items": { "type": "string" } }
        }
      }
    }
  }
}

Usage Example

javascript
const feedbackAnalysis = await executeSkill('feedback-aggregation', {
  sources: [
    {
      type: 'support-tickets',
      data: supportTickets,
      dateRange: { start: '2026-01-01', end: '2026-01-24' }
    },
    {
      type: 'nps-verbatim',
      data: npsResponses
    },
    {
      type: 'app-reviews',
      data: appStoreReviews
    }
  ],
  analysisScope: 'all',
  segmentation: ['plan_type', 'company_size', 'tenure']
});

Dependencies

NLP capabilities
Support platform APIs (Intercom, Zendesk)
App store APIs

Illustrative Want to see what a5c-ai/Customer Feedback Aggregation 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, 21:20 UTC · 24 cases

Lift

+33%

79% vs 46%

Agent turns

0%

1.0 vs 1.0

Tokens

+37%

2,726.46 vs 1,995.13

fail → pass · 9pass → pass · 10fail → fail · 4pass → fail · 1

Per-case results (24)

Case	Status	Duration (ms)			Turns			Tokens			Tool calls
Case	Status	Without	With	Δ	Without	With	Δ	Without	With	Δ	Without	With	Δ
case-01 We just finished Q1 and I need to synthesize our customer voice data from recent Intercom support tickets and NPS verbatim survey responses. Please aggregate these inputs to provide a summary detailing total item volume and overall sentiment, recurring feedback themes with example quotes, a detailed list of feature requests showing request counts and urgency scores, critical pain points rated by severity, and prioritized recommendations backed by customer evidence.	fail→fail	27,215	32,268	+19%	1	1	0%	4,590	5,894	+28%	0	0	—
case-02 I am analyzing sales call notes and app store reviews from the last 30 days, segmented by SMB vs enterprise accounts. Please process this data and output a structured analysis that includes source volume breakdowns, overall sentiment trends, emerging and declining feedback topics, pain points categorized by customer impact, and feature requests mapped across account segments with representative quotes.	fail→fail	21,934	27,406	+25%	1	1	0%	3,756	5,932	+58%	0	0	—
case-03 Our product team completed 25 customer interviews and collected 200 survey responses about our reporting module. Could you process these qualitative inputs and generate a comprehensive synthesis report containing high-level volume metrics, themes segmented by user plan type, a list of requested features with impact estimates and quote references, and action-oriented recommendations prioritized by impact?	fail→fail	28,147	29,741	+6%	1	1	0%	4,921	7,255	+47%	0	0	—
case-04 We have collected customer feedback metrics and need to calculate the RICE scores for three candidate features: Automated Invoicing (Reach: 5000, Impact: 3, Confidence: 80%, Effort: 2 person-months), Single Sign-On (Reach: 1000, Impact: 3, Confidence: 100%, Effort: 1 person-month), and Export to CSV (Reach: 8000, Impact: 1, Confidence: 90%, Effort: 0.5 person-months). Please output the calculated RICE scores and rank the features accordingly.	pass→pass	6,634	7,009	+6%	1	1	0%	1,753	3,025	+73%	0	0	—
case-05 Based on customer feedback calling for better export tools, write a detailed Product Requirements Document (PRD) section specifying the functional requirements, acceptance criteria, and user flow for a CSV Data Export feature in our web application.	pass→pass	19,006	24,172	+27%	1	1	0%	3,617	5,631	+56%	0	0	—
case-06 We are hosting our quarterly Customer Advisory Board meeting next month. Create a 90-minute meeting agenda and discussion guide to engage enterprise executives on our product roadmap.	pass→fail	17,323	21,819	+26%	1	1	0%	2,763	4,959	+79%	0	0	—
case-07 When assembling an input JSON object for customer feedback processing, which top-level field is marked as required in the schema?	fail→pass	3,600	1,510	-58%	1	1	0%	504	1,692	+236%	0	0	—
case-08 We are configuring an input payload for feedback processing across six data sources: support tickets, NPS verbatim surveys, sales calls, app store reviews, customer interviews, and general surveys. What exact hyphenated enum string values does the input schema specify for these six source types?	pass→pass	7,408	2,879	-61%	1	1	0%	1,376	2,060	+50%	0	0	—
case-09 In the input schema for customer feedback aggregation, what are the allowed enum string values for the analysisScope property when selecting focus areas such as analyzing everything, feature requests, pain points, sentiment, or trends?	pass→pass	7,496	1,898	-75%	1	1	0%	1,392	1,760	+26%	0	0	—
case-10 What key fields comprise the top-level summary object in the feedback aggregation JSON output schema?	fail→pass	11,817	32,396	+174%	1	1	0%	2,305	1,826	-21%	0	0	—
case-11 What fields are defined for each individual theme object inside the top-level themes array in the output JSON schema?	fail→fail	6,163	2,725	-56%	1	1	0%	1,002	1,942	+94%	0	0	—
case-12 In the output JSON schema for feedback aggregation, what specific property name is used to record the numerical urgency rating of a feature request?	fail→pass	7,468	2,171	-71%	1	1	0%	1,305	1,860	+43%	0	0	—
case-13 What fields make up each object in the painPoints array of the output JSON schema?	fail→pass	7,148	2,223	-69%	1	1	0%	1,051	1,740	+66%	0	0	—
case-14 How is the trends object structured in the output JSON schema for feedback aggregation?	pass→pass	16,066	3,300	-79%	1	1	0%	2,740	2,060	-25%	0	0	—
case-15 In the output JSON schema, what field inside each recommendation item holds supporting quotes or data points?	pass→pass	5,840	1,604	-73%	1	1	0%	990	1,718	+74%	0	0	—
case-16 Which target process JavaScript files are designated for integration with feedback aggregation?	fail→pass	11,687	2,394	-80%	1	1	0%	1,815	1,807	-0%	0	0	—
case-17 In the input payload JSON schema, how is the timeRange object defined and what date format is required for start and end?	pass→pass	6,147	3,250	-47%	1	1	0%	1,179	1,983	+68%	0	0	—
case-18 When configuring an input payload to categorize feedback by dimensions like plan_type or company_size, under which top-level schema key is this array passed?	fail→pass	11,141	3,358	-70%	1	1	0%	1,922	1,929	+0%	0	0	—
case-19 What are the six top-level keys required by the JSON output schema for feedback aggregation?	fail→pass	12,326	2,033	-84%	1	1	0%	2,149	1,776	-17%	0	0	—
case-20 What exact skill identifier string is passed to executeSkill in the JavaScript usage code example?	fail→pass	3,263	1,620	-50%	1	1	0%	506	1,717	+239%	0	0	—
case-21 What technical dependencies are specified for customer feedback aggregation?	pass→pass	10,183	1,503	-85%	1	1	0%	1,811	1,679	-7%	0	0	—
case-22 In the output JSON schema, what is the exact property name used inside featureRequests items to store sample customer quotes?	fail→pass	10,799	2,050	-81%	1	1	0%	1,779	1,799	+1%	0	0	—
case-23 What data collection sources are listed under capabilities for synthesizing voice of customer inputs?	pass→pass	8,528	7,113	-17%	1	1	0%	1,443	1,685	+17%	0	0	—
case-24 What synthesis deliverables are listed under the capabilities overview for customer feedback synthesis?	pass→pass	7,868	1,609	-80%	1	1	0%	1,214	1,706	+41%	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. 24 cases were attempted. The headline lift of +33 percentage points is the difference between those two pass rates over the 24 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.

Model measured: gemini-3.6-flash
Graded by: gemini-3.6-flash
Grading method: judged
Verification: verified
Pass rate without the skill: 45.8%
Pass rate with the skill: 79.2%
Cases attempted: 24
Cases counted toward lift: 24
Cases that regressed: 1
Skill version measured: v1
Run completed: 8/29/2026, 21:20 UTC

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

Prompt

When assembling an input JSON object for customer feedback processing, which top-level field is marked as required in the schema?

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 run79

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+33% 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.