Data Insight Narrator
Turns raw metrics and query results into decision-ready narratives: findings, so-what, confidence and recommended actions.
Overview
Data Insight Narrator is the bridge between a spreadsheet of numbers and a room of decision-makers. Feed it query results, dashboards exports or summary metrics and it produces an executive-ready narrative: what changed, why it matters, how confident we are, and what to do next.
It separates observation from interpretation, quantifies impact in business terms, flags anomalies and seasonality instead of over-explaining noise, and explicitly lists what the data does NOT show.
The output format works for weekly business reviews, product analytics digests and investor updates alike.
System prompt
Preview redactedThe instruction block that defines the agent's role, rules and output format.
████████████████████████████████████████████████████████████████
████████████████████████████████████████████████████████████████
████████████████████████████████████████████████████████████████
████████████████████████████████████████████████████████████████
████████████████████████████████████████████████████████████████
████████████████████████████████████████████████████████████████
████████████████████████████████████████████████████████████████
████████████████████████████████████████████████████████████████
████████████████████████████████████████████████████████████████
████████████████████████████████████████████████████████████████
████████████████████████████████████████████████████████████████
████████████████████████████████████████████████████████████████
████████████████████████████████████████████████████████████████
████████████████████████████████████████████████████████████████
████████████████████████████████████████████████████████████████
████████████████████████████████████████████████████████████████
████████████████████████████████████████████████████████████████
████████████████████████████████████████████████████████████████
████████████████████████████████████████████████████████████████
████████████████████████████████████████████████████████████████
████████████████████████████████████████████████████████████████
████████████████████████████████████████████████████████████████ User prompt template
Fill in the variables and send it to run the skill: {{variables}}
████████████████████████████████████████████████████████████████
████████████████████████████████████████████████████████████████
████████████████████████████████████████████████████████████████
████████████████████████████████████████████████████████████████
████████████████████████████████████████████████████████████████
████████████████████████████████████████████████████████████████
████████████████████████████████████████████████████████████████
████████████████████████████████████████████████████████████████
████████████████████████████████████████████████████████████████
████████████████████████████████████████████████████████████████
████████████████████████████████████████████████████████████████
████████████████████████████████████████████████████████████████ Variables
| Name | Required | Description | Example |
|---|---|---|---|
| {{metrics}} | Yes | Raw numbers, CSV/JSON results or dashboard summary. | — |
| {{dataset_desc}} | Yes | What the data covers and how it was collected. | Postgres events table, signup-to-activation funnel, last 90 days |
| {{time_window}} | Optional | The reporting period and comparison baseline. | Sep 1–30 vs Aug, UTC |
| {{audience}} | Optional | Who reads the narrative. | exec team at a seed-stage SaaS |
Full skill configuration
Redacted previewThe standard .aiskill file in JSON and YAML.
{
"format": "aiskill/v1",
"skill": {
"name": "Data Insight Narrator",
"slug": "data-insight-narrator",
"version": "1.2.0",
"description": "Turns raw metrics and query results into decision-ready narratives: findings, so-what, confidence and recommended actions.",
"category": "Data",
"tags": [
"analytics",
"reporting",
"storytelling",
"business-intelligence"
],
"platforms": [
"chatgpt",
"dify",
"langgraph"
],
"tested_environment": "GPT-5, Claude Sonnet 4.5; CSV/JSON exports up to ~5k rows summarized",
"license": "CC-BY-4.0"
},
"prompts": {
"system": "████████████████████████████████████████████████████████████████████████████████████████████████",
"user_template": "████████████████████████████████████████████████████████████████████████████████████████████████"
},
"variables": [
{
"name": "metrics",
"description": "Raw numbers, CSV/JSON results or dashboard summary.",
"required": true
},
{
"name": "dataset_desc",
"description": "What the data covers and how it was collected.",
"required": true,
"example": "Postgres events table, signup-to-activation funnel, last 90 days"
},
{
"name": "time_window",
"description": "The reporting period and comparison baseline.",
"required": false,
"example": "Sep 1–30 vs Aug, UTC"
},
{
"name": "audience",
"description": "Who reads the narrative.",
"required": false,
"example": "exec team at a seed-stage SaaS"
}
],
"install": {
"steps": [
{
"title": "Copy the system prompt",
"detail": "Copy the narrator system prompt from the skill page."
},
{
"title": "Create a reporting agent",
"detail": "Paste it into a custom GPT or Dify/LangGraph reporting agent."
},
{
"title": "Paste your numbers",
"detail": "Provide the metrics, dataset context and time window per report."
},
{
"title": "Distribute",
"detail": "Drop the Headline + Recommended actions into your weekly review doc."
}
]
},
"usage": {
"steps": [
"Export the metrics you want narrated, including a comparison baseline.",
"Fill in dataset context so confidence ratings are meaningful.",
"Review hypothesis labels and verify any HIGH-impact claim before acting.",
"Track the Watch list metrics in your next reporting cycle."
]
},
"metadata": {
"author": "SkillVault",
"source": "https://skillvault.top/skills/data-insight-narrator",
"exported_at": "2026-11-03T07:43:38.894Z"
}
} format: aiskill/v1
skill:
name: Data Insight Narrator
slug: data-insight-narrator
version: 1.2.0
description: >-
Turns raw metrics and query results into decision-ready narratives: findings, so-what,
confidence and recommended actions.
category: Data
tags:
- analytics
- reporting
- storytelling
- business-intelligence
platforms:
- chatgpt
- dify
- langgraph
tested_environment: GPT-5, Claude Sonnet 4.5; CSV/JSON exports up to ~5k rows summarized
license: CC-BY-4.0
prompts:
system: ████████████████████████████████████████████████████████████████████████████████████████████████
user_template: ████████████████████████████████████████████████████████████████████████████████████████████████
variables:
- name: metrics
description: Raw numbers, CSV/JSON results or dashboard summary.
required: true
- name: dataset_desc
description: What the data covers and how it was collected.
required: true
example: Postgres events table, signup-to-activation funnel, last 90 days
- name: time_window
description: The reporting period and comparison baseline.
required: false
example: Sep 1–30 vs Aug, UTC
- name: audience
description: Who reads the narrative.
required: false
example: exec team at a seed-stage SaaS
install:
steps:
- title: Copy the system prompt
detail: Copy the narrator system prompt from the skill page.
- title: Create a reporting agent
detail: Paste it into a custom GPT or Dify/LangGraph reporting agent.
- title: Paste your numbers
detail: Provide the metrics, dataset context and time window per report.
- title: Distribute
detail: Drop the Headline + Recommended actions into your weekly review doc.
usage:
steps:
- Export the metrics you want narrated, including a comparison baseline.
- Fill in dataset context so confidence ratings are meaningful.
- Review hypothesis labels and verify any HIGH-impact claim before acting.
- Track the Watch list metrics in your next reporting cycle.
metadata:
author: SkillVault
source: https://skillvault.top/skills/data-insight-narrator
exported_at: "2026-11-03T07:43:38.894Z"
Installation guide
- 1
Copy the system prompt
Copy the narrator system prompt from the skill page.
- 2
Create a reporting agent
Paste it into a custom GPT or Dify/LangGraph reporting agent.
- 3
Paste your numbers
Provide the metrics, dataset context and time window per report.
- 4
Distribute
Drop the Headline + Recommended actions into your weekly review doc.
How to use
- Export the metrics you want narrated, including a comparison baseline.
- Fill in dataset context so confidence ratings are meaningful.
- Review hypothesis labels and verify any HIGH-impact claim before acting.
- Track the Watch list metrics in your next reporting cycle.
Platform compatibility
ChatGPT GPTs
OpenAI
Paste the system prompt into the GPT "Instructions" field and keep the user template as your starter message.
Official docsDify
LangGenius
Create a new Agent/App, paste the system prompt as the system prompt, and register template variables as input fields.
Official docsLangGraph
LangChain
Load the JSON config into your graph state: use prompts.system for the system message and prompts.user_template as a runnable template.
Official docsPreview gallery

Related skills
SQL Query Builder
Translates plain-English questions into safe, indexed, dialect-aware SQL — with schema reasoning and EXPLAIN prompts.
Landing Page Architect
Designs high-converting landing pages end-to-end: section wireframe, full copy draft, CTA strategy and A/B test plan. Yearly Pro exclusive.
Deep Research Agent
A disciplined research protocol: question decomposition, source strategy, evidence table and a confidence-rated synthesis.