AgentSkillsCN

Explore

探索

SKILL.md

Skill: Explore Data

Purpose

Quick, interactive data exploration without the full pipeline. Lets users poke around the active dataset — preview tables, check distributions, spot patterns, and form hypotheses before committing to a formal analysis.

When to Use

  • User says /explore or "let me explore the data" or "what's in this dataset?"
  • After connecting a new dataset, before any formal analysis
  • When the user wants to understand data shape without a specific question

Invocation

/explore — explore the active dataset /explore {table} — focus on a specific table /explore {table} {column} — deep-dive into a specific column

Instructions

Step 1: Load Context

Read .knowledge/active.yaml to identify the active dataset. Read .knowledge/datasets/{active}/schema.md for table/column reference. Read .knowledge/datasets/{active}/quirks.md for known gotchas.

If no active dataset, prompt: "No dataset connected. Use /connect-data to add one."

Step 2: Choose Exploration Mode

Mode A: Dataset overview (no table specified)

  • List all tables with row counts and date ranges
  • Highlight the 3-5 most analytically useful tables (most rows, most joins)
  • Show key entities and how they connect
  • Suggest 3 starting questions based on available data

Mode B: Table exploration (table specified)

  • Show column list with types and null rates
  • Sample 5 random rows
  • For numeric columns: min, max, mean, median
  • For categorical columns: top 5 values with counts
  • For date columns: range and coverage
  • Flag any quality issues (>5% nulls, low cardinality, suspicious values)

Mode C: Column deep-dive (table + column specified)

  • Full distribution: histogram for numeric, bar chart for categorical
  • Null analysis: count, pattern (random vs systematic)
  • Outlier detection: IQR method, flag extremes
  • If date column: coverage heatmap by week
  • Suggest related columns for cross-analysis

Step 3: Interactive Follow-Up

After presenting results, offer 2-3 contextual next actions:

  • "Want to see how {column} varies by {dimension}?"
  • "This looks like a good candidate for funnel analysis. Want to try /run-pipeline?"
  • "There are quality issues in {column}. Want to run /data-profiling?"

Step 4: Save Exploration Notes

Write a brief exploration summary to working/explore_notes_{DATE}.md:

  • Tables examined
  • Key observations
  • Quality flags
  • Suggested next steps

This file is available for subsequent agents (e.g., Question Framing can reference exploration notes to inform hypothesis generation).

Rules

  1. Keep it fast — no more than 3-4 queries per exploration step
  2. Always apply swd_style() if generating any chart
  3. Never modify data during exploration
  4. Always cite table and column names in output
  5. If data source is CSV fallback, mention this to the user

Edge Cases

  • Empty table: Report row count = 0, suggest checking data load
  • Table not found: Fuzzy-match against schema, suggest closest match
  • Column has all nulls: Flag as BLOCKER, suggest checking data pipeline
  • Very wide table (>50 columns): Group columns by category, show summary not full list