JPeetz/agent-skills / skills/data/data-governance
data-governance
VerifiedUse this skill when establishing or auditing data governance practices — data cataloging and discovery, data lineage tracking, data quality frameworks, metadata management, data classification and tagging, ownership and stewardship models, data retention policies, and compliance mapping (GDPR, CCPA, SOC 2 data controls). Covers tools like Atlan, DataHub, Amundsen, dbt, Great Expectations, and Soda.
| File | Kind | Size |
|---|---|---|
| CHANGELOG.md | markdown | 2.2 KB |
| LICENSE | other | 1.1 KB |
| SKILL.md | skill md | 25.0 KB |
| evals/evals.json | config | 3.9 KB |
| references/sources.md | markdown | 5.1 KB |
| scripts/scan_governance.py | script | 7.9 KB |
evals/evals.json
View on GitHub[
{
"id": "gov-001",
"description": "User asks to stand up a governance program from scratch",
"input": "We have a Snowflake warehouse plus dbt and about 400 tables, no governance at all. Where do we start?",
"expected_trigger": true,
"expected_output": "A phased plan that starts with people and decisions, not a tool: identify systems and 10 tier-1 datasets, name a sponsor/governance lead, add owner/domain/tier/classification/retention meta to those 10 in dbt, then stand up a catalog and ingest dbt + warehouse metadata, publish a 6-page policy (scope, roles, 4-tier classification, SLOs, retention, exception process), add a metadata CI gate on changed files, and run the first quarterly rituals. Explicitly rejects 'buy a catalog first' and 'govern all 400 tables'.",
"assertions": [
"plan is phased (baseline -> decisions -> instrumentation -> operate) with timeboxes",
"starts from tier-1 datasets, not full inventory",
"includes owners, classification tiers (max 4), SLOs, retention, and an expiring exception process",
"metadata is defined in dbt meta / version control and pushed to the catalog",
"includes at least one CI gate and a recurring cadence",
"does not recommend catalog purchase as step 1"
]
},
{
"id": "gov-002",
"description": "Column-level lineage impact analysis before a breaking change",
"input": "I need to rename dim_customers.email to email_hash and change its type. What's downstream and how do I not break the exec dashboard?",
"expected_trigger": true,
"expected_output": "A column-level impact analysis: locate the asset, expand downstream 2-3 hops filtered to material consumers (tier-1 dashboards, ML features, reverse-ETL, exports), notify downstream owners with a deadline, prefer an additive shim (keep email, add email_hash, set meta.deprecation_date) over a hard rename, verify type change compatibility (string widening, hash format), and record consumers/notification/deprecation date in the PR. Notes that table-level lineage alone is insufficient and that a classification tag must follow the column to the new name (PII + masking).",
"assertions": [
"uses column-level lineage, not just table-level",
"filters to material/tier-1 downstream consumers and identifies their owners",
"recommends additive migration with a deprecation date instead of a hard rename",
"carries pii + masking metadata to the renamed column",
"documents the impact set and notification in the change record"
]
},
{
"id": "gov-003",
"description": "Compliance mapping for GDPR and SOC 2 data controls",
"input": "We're doing SOC 2 Type II and we also get GDPR data subject requests. What governance controls and evidence do we need?",
"expected_trigger": true,
"expected_output": "A control matrix mapping each requirement to a concrete artifact and the evidence it produces: GDPR Art. 30 processing register generated from the catalog, Art. 6 lawful_basis tag coverage, Art. 15-17 DSR runbook with a subject index and ticket samples, Art. 32 masking/access enforcement coverage, Art. 33 72-hour breach process scoped by lineage; SOC 2 CC6.1/CC6.3 access matrix plus quarterly recertification, CC6.7 retention and purge with a deletion log, CC7.2/CC7.4 monitoring alerts and closed incidents, P4.1-P4.3 retention schedule and purge evidence. Recommends one crosswalk table (control_id -> clause -> artifact path -> owner -> last verified) and dated automated exports over screenshots, retained for the audit window.",
"assertions": [
"maps requirements to artifacts AND to the evidence produced",
"covers GDPR Arts. 6, 15-17, 30, 32, 33 and SOC 2 CC6/CC7/P4",
"includes retention/purge evidence (deletion log) as a shared control",
"recommends a reusable crosswalk rather than per-framework documentation",
"specifies evidence hygiene (dated exports, 12-month retention for Type II)"
]
}
]
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| rea-analysis-optimizationmorluto/reaOptimize REA providers and shared analysis workflows for agent context, latency, CPU, and memory while preserving correctness and Evidence. Use for analysis bottlenecks and cache or query design; CI/test-suite optimization belongs to rea-ci-tests-audit. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| rea-ci-tests-auditmorluto/reaAudit REA CI and test suites for correctness, fixture drift, false passes, and measured performance; use for suite or workflow investigations and efficiency changes, rather than ordinary feature testing or a single test failure. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| rea-tool-designmorluto/reaAudit, investigate, design, or change REA investigation tools, CLI/MCP contracts, provider capabilities, and Evidence semantics. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-auditDietrichGebert/ponytailQuality audit of the whole repo: bugs, security, real load, missing tests, speed, and what to delete. Most important first. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |