Skip to content

JPeetz/agent-skills / skills/data/exploratory-data-analysis

exploratory-data-analysis

Verified

Use this skill when exploring an unfamiliar dataset, investigating data quality issues, validating data pipeline outputs before production use, reviewing SQL queries for correctness and performance, or producing a data quality scorecard for a data asset.

FileKindSize
CHANGELOG.mdmarkdown757 B
LICENSEother979 B
SKILL.mdskill md8.7 KB
evals/evals.jsonconfig3.3 KB

evals/evals.json

View on GitHub
[
  {
    "id": "eda-001",
    "description": "First exploration of an unfamiliar dataset",
    "input": "I just got this orders.parquet from the new vendor — explore it before we build anything on it.",
    "shouldTrigger": true,
    "expect": "Agent loads and profiles the data: shape, columns, dtypes, describe() for numeric and object columns, then null analysis, duplicates, and distributions.",
    "assertions": [
      "data profile reports shape, columns, and dtypes",
      "null profile is computed with counts and percentages",
      "exact and key-based duplicates are checked"
    ]
  },
  {
    "id": "eda-002",
    "description": "Data quality audit with freshness check",
    "input": "Validate the pipeline output before we consume it in the dashboard.",
    "shouldTrigger": true,
    "expect": "Agent runs a data quality audit: freshness check against the SLA, value range validation against domain rules, referential integrity, and a quality scorecard.",
    "assertions": [
      "freshness/staleness is compared to expected lag",
      "validation rules are applied per column",
      "a quality scorecard with dimension scores and evidence is produced"
    ]
  },
  {
    "id": "eda-003",
    "description": "Root cause analysis of a metric drop",
    "input": "Active users dropped 40% this week — what happened?",
    "shouldTrigger": true,
    "expect": "Agent follows the root cause workflow: verify the data/pipeline, check the denominator, segment by dimension, look for correlated changes, check data quality, compare to prior periods, and document findings.",
    "assertions": [
      "data/pipeline health is verified first",
      "the metric denominator is checked",
      "findings are documented with what was checked, found, and still uncertain"
    ]
  },
  {
    "id": "eda-004",
    "description": "SQL correctness and performance review",
    "input": "Review this SQL query for correctness and performance before we ship it.",
    "shouldTrigger": true,
    "expect": "Agent validates correctness (joins, aggregations, NULL handling, window scoping), performance (scans, index usage, SELECT *), and security (parameterization) against the checklist.",
    "assertions": [
      "JOIN conditions are checked for accidental Cartesian products",
      "sargable filters and index usage are reviewed",
      "query parameterization is verified"
    ]
  },
  {
    "id": "eda-005",
    "description": "Produce a formal data quality scorecard",
    "input": "Produce a data quality scorecard for the customer_dimension asset.",
    "shouldTrigger": true,
    "expect": "Agent scores completeness, consistency, uniqueness, timeliness, accuracy, and integrity dimensions (1-5) with evidence for each.",
    "assertions": [
      "all six scorecard dimensions are rated",
      "each score carries evidence",
      "findings include severity ratings"
    ]
  },
  {
    "id": "eda-006",
    "description": "Near-miss negative: automatic fixing instead of diagnosis",
    "input": "Fix the nulls and duplicate rows in this dataset for me.",
    "shouldTrigger": false,
    "expect": "EDA is diagnostic — it does not fix data quality issues automatically; cleaning belongs to data-analysis skills.",
    "assertions": [
      "skill does not trigger on requests to directly rewrite/fix data"
    ]
  }
]
writing-skillsobra/superpowersUse when creating new skills, editing existing skills, or verifying skills work before deploymentStars: 322kUpdated Claude CodeCodexVerifiedInstall
xlsxanthropics/skillsUse this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .xltx, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like "the xlsx in my downloads") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.Stars: 198kUpdated Claude CodeCodexVerifiedInstall
shipping-and-launchaddyosmani/agent-skillsPrepares production launches. Use when preparing to deploy to production, or when asking what needs to be in place before shipping. Use when you need a pre-launch checklist, when setting up monitoring, when planning a staged rollout, or when you need a rollback strategy.Stars: 119kUpdated Claude CodeCodexVerifiedInstall
childhoodthedotmack/claude-memBabysit a shipped release after merge, publication, or deployment. Verify published packages, production behavior, installed plugins, and late CI or user reports until the observation window is clean; use babysit for PR readiness before release.Stars: 118kUpdated Claude CodeCodexVerifiedInstall