AgentSkillsCN

validation-history-integration

利用 store_validation_result() 集成工具，自动存储验证结果。

SKILL.md
--- frontmatter
name: validation-history-integration
description: Auto-store validation results using store_validation_result() integration utility
license: MIT
compatibility: opencode,claude
metadata:
  created_at: "2026-01-29T22:58:59.652Z"
  updated_at: "2026-01-29T22:58:59.652Z"
  version: "1"
  tags: "validation,history,integration"
  compatibility: "2.1.0+"
  difficulty: "intermediate"
<!-- BEGIN:compound:skill-managed -->

Validation History Integration

This skill provides guidance on automatically storing validation results in history.

When To Use

  • •User asks about auto-storing validation results
  • •Implementing validation framework with history persistence
  • •Need to integrate ValidationSuite with ValidationHistoryStore

Procedure

  1. •Import from vibe_piper.validation.integration: store_validation_result
  2. •Create ValidationHistoryStore (PostgreSQLValidationHistoryStore)
  3. •Call store.initialize_schema() before first use
  4. •Run validation suite on data
  5. •Call store_validation_result() with:
    • •SuiteValidationResult from suite.validate()
    • •Asset name (string)
    • •History store instance
    • •Optional pipeline_id for tracking
  6. •Returns validation_run_id for querying history later

Example

python
from vibe_piper.validation import ValidationSuite, expect_column_values_to_be_unique
from vibe_piper.validation.integration import store_validation_result
from vibe_piper.connectors.postgres import PostgreSQLConnector, PostgreSQLConfig

suite = ValidationSuite(name='data_quality')
suite.add_check('unique_ids', expect_column_values_to_be_unique('user_id'))

config = PostgreSQLConfig(host='localhost', port=5432, database='vp', user='u', password='p')
connector = PostgreSQLConnector(config)
connector.connect()

store = PostgreSQLValidationHistoryStore(connector)
store.initialize_schema()

result = suite.validate(data)
run_id = store_validation_result(result, asset_name='users', history_store=store, pipeline_id='daily')
print(f'Stored as: {run_id}')

Integration Points

  • •Works with existing ValidationSuite API
  • •Works with all validation check functions
  • •Integrates with asset metadata (pipeline_id, suite_name)
  • •Stores comprehensive metrics for trend analysis

Notes

  • •Requires PostgreSQL database initialized with validation history schema
  • •Call store.initialize_schema() before first use
  • •Metrics are automatically extracted: pass_rate, duration_ms, total_records
  • •Validation run ID can be used for tracking and querying history
  • •See src/vibe_piper/validation/history.py for full API
<!-- END:compound:skill-managed -->

Manual notes

This section is preserved when the skill is updated. Put human notes, caveats, and exceptions here.