skills.rest
diankong720-ui
diankong720-uiCOMMUNITY
Explore all skills by this author →
GitHub

data-discovery

Probe warehouse environments and map observed facts into a DataContextBundle.

Updated Aug 27, 2026
Search Tags:#data-discovery#stage-2#data-context#llm-runtime#warehouse-facts#data-context-bundle
One-click install
npx skills add https://github.com/diankong720-ui/Pandora --skill data-discovery
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: data-discovery
Source: https://github.com/diankong720-ui/Pandora/tree/main/skills/data-discovery
Command: npx skills add https://github.com/diankong720-ui/Pandora --skill data-discovery

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

The Data Discovery skill helps the runtime collect warehouse facts and interpret them into a DataContextBundle for downstream planning.

Core Features & Use Cases

  • Probes the environment to gather warehouse facts such as visible tables, header information, sample rows, and current warehouse load status.
  • Maps raw findings into the shared DataContextBundle fields (environment_scan, schema_map, metric_mapping, time_fields, dimension_fields, supported_dimension_capabilities, joinability, comparison_feasibility, warehouse_load_status, etc.).
  • Produces discovery-time risk signals via report_conflict_hint when critical ambiguities are detected and ensures the data is used strictly for discovery, not for claim validation or decision-making.

Quick Start

Run a discovery session to produce a DataContextBundle from the current warehouse environment.

Frequently Asked Questions about data-discovery

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I map warehouse facts into a structured data context for planning?▼

Mapping warehouse facts involves probing visible tables, headers, and sample rows to populate a DataContextBundle with environment scans and schema maps for downstream planning.

What is a DataContextBundle and how does it support data discovery?▼

A DataContextBundle is a structured output containing schema maps, metric mappings, and time fields derived strictly from observed warehouse facts to support Stage 2 planning workflows.

How do I generate schema maps and metric mappings from my warehouse environment?▼

Generate schema maps and metric mappings by running a discovery session that probes the current warehouse environment and translates raw findings into structured context fields.

Can I use warehouse facts to validate claims or make prescriptive decisions?▼

No, warehouse facts must be used strictly for discovery. The process avoids prescriptive conclusions and instead exposes explicit data-quality signals and conflict hints for evaluation.

What warehouse load status and joinability signals do I need for comparison feasibility?▼

Comparison feasibility requires warehouse load status, joinability, and supported dimension capabilities signals derived from observed facts to evaluate downstream planning risks.

Are there limitations when deriving metric mappings from ambiguous warehouse facts?▼

Yes, critical ambiguities trigger discovery-time risk signals via conflict hints, limiting prescriptive conclusions and ensuring data is used only for context, not claim validation.

Related Skills

Explore More Data & Analytics Skills→
lollipopkitlollipopkit
1

supabase-rest

Inspect Supabase tables, columns, and sample rows via read-only REST calls.

Community
Intermediate
ChegginCheggin
1

deep-dive

Run a two-stage trace and interview workflow to produce a testable specification.

Community
Advanced
DamarisDengDamarisDeng

score-and-rank

Score and rank research questions by blending data feasibility with literature signals.

Community
Advanced
auldsyababuaauldsyababua
6

resume-manager

Extract resume data into a structured database and generate tailored PDFs.

Community
Advanced