Skills › Knowledge Work Plugins › data-context-extractor › skill-template.md

skill-template.md

148 lines · data/skills/data-context-extractor/references/skill-template.md at 2574e29

Open on GitHub

1# Generated Skill Template

3Use this template when generating a new data analysis skill. Replace all `[PLACEHOLDER]` values.

5---

7```markdown

8---

9name: [company]-data-analyst

10description: "[COMPANY] data analysis skill. Provides context for querying [WAREHOUSE_TYPE] including entity definitions, metric calculations, and common query patterns. Use when analyzing [COMPANY] data for: (1) [PRIMARY_USE_CASE_1], (2) [PRIMARY_USE_CASE_2], (3) [PRIMARY_USE_CASE_3], or any data questions requiring [COMPANY]-specific context."

11---

13# [COMPANY] Data Analysis

15## SQL Dialect: [WAREHOUSE_TYPE]

17[INSERT APPROPRIATE DIALECT SECTION FROM sql-dialects.md]

19---

21## Entity Disambiguation

23When users mention these terms, clarify which entity they mean:

25[EXAMPLE FORMAT - customize based on discovery:]

27**"User" can mean:**

28- **Account**: An individual login/profile ([PRIMARY_TABLE]: [ID_FIELD])

29- **Organization**: A billing entity that can have multiple accounts ([ORG_TABLE]: [ORG_ID])

30- **[OTHER_TYPE]**: [DEFINITION] ([TABLE]: [ID])

32**Relationships:**

33- [ENTITY_1] → [ENTITY_2]: [RELATIONSHIP_TYPE] (join on [JOIN_KEY])

35---

37## Business Terminology

39| Term | Definition | Notes |

40|------|------------|-------|

41| [TERM_1] | [DEFINITION] | [CONTEXT/GOTCHA] |

42| [TERM_2] | [DEFINITION] | [CONTEXT/GOTCHA] |

43| [ACRONYM] | [FULL_NAME] - [EXPLANATION] | |

45---

47## Standard Filters

49Always apply these filters unless explicitly told otherwise:

51```sql

52-- Exclude test/internal data

53WHERE [TEST_FLAG_COLUMN] = FALSE

54 AND [INTERNAL_FLAG_COLUMN] = FALSE

56-- Exclude invalid/fraud

57 AND [STATUS_COLUMN] != '[EXCLUDED_STATUS]'

59-- [OTHER STANDARD EXCLUSIONS]

60```

62**When to override:**

63- [SCENARIO_1]: Include [NORMALLY_EXCLUDED] when [CONDITION]

65---

67## Key Metrics

69### [METRIC_1_NAME]

70- **Definition**: [PLAIN_ENGLISH_EXPLANATION]

71- **Formula**: `[EXACT_CALCULATION]`

72- **Source**: `[TABLE_NAME].[COLUMN_NAME]`

73- **Time grain**: [DAILY/WEEKLY/MONTHLY]

74- **Caveats**: [EDGE_CASES_OR_GOTCHAS]

76### [METRIC_2_NAME]

77[REPEAT FORMAT]

79---

81## Data Freshness

83| Table | Update Frequency | Typical Lag |

84|-------|------------------|-------------|

85| [TABLE_1] | [FREQUENCY] | [LAG] |

86| [TABLE_2] | [FREQUENCY] | [LAG] |

88To check data freshness:

89```sql

90SELECT MAX([DATE_COLUMN]) as latest_data FROM [TABLE]

91```

93---

95## Knowledge Base Navigation

97Use these reference files for detailed table documentation:

99| Domain | Reference File | Use For |

100|--------|----------------|---------|

101| [DOMAIN_1] | `references/[domain1].md` | [BRIEF_DESCRIPTION] |

102| [DOMAIN_2] | `references/[domain2].md` | [BRIEF_DESCRIPTION] |

103| Entities | `references/entities.md` | Entity definitions and relationships |

104| Metrics | `references/metrics.md` | KPI calculations and formulas |

105

106---

107

108## Common Query Patterns

109

110### [PATTERN_1_NAME]

111```sql

112[SAMPLE_QUERY]

113```

114

115### [PATTERN_2_NAME]

116```sql

117[SAMPLE_QUERY]

118```

119

120---

121

122## Troubleshooting

123

124### Common Mistakes

125- **[MISTAKE_1]**: [EXPLANATION] → [CORRECT_APPROACH]

126- **[MISTAKE_2]**: [EXPLANATION] → [CORRECT_APPROACH]

127

128### Access Issues

129- If you encounter permission errors on `[TABLE]`: [WORKAROUND]

130- For PII-restricted columns: [ALTERNATIVE_APPROACH]

131

132### Performance Tips

133- Filter by `[PARTITION_COLUMN]` first to reduce data scanned

134- For large tables, use `LIMIT` during exploration

135- Prefer `[AGGREGATED_TABLE]` over `[RAW_TABLE]` when possible

136```

137

138---

139

Generated Skill Template

Use this template when generating a new data analysis skill. Replace all [PLACEHOLDER] values.

---
name: [company]-data-analyst
description: "[COMPANY] data analysis skill. Provides context for querying [WAREHOUSE_TYPE] including entity definitions, metric calculations, and common query patterns. Use when analyzing [COMPANY] data for: (1) [PRIMARY_USE_CASE_1], (2) [PRIMARY_USE_CASE_2], (3) [PRIMARY_USE_CASE_3], or any data questions requiring [COMPANY]-specific context."
---

# [COMPANY] Data Analysis

## SQL Dialect: [WAREHOUSE_TYPE]

[INSERT APPROPRIATE DIALECT SECTION FROM sql-dialects.md]

---

## Entity Disambiguation

When users mention these terms, clarify which entity they mean:

[EXAMPLE FORMAT - customize based on discovery:]

**"User" can mean:**
- **Account**: An individual login/profile ([PRIMARY_TABLE]: [ID_FIELD])
- **Organization**: A billing entity that can have multiple accounts ([ORG_TABLE]: [ORG_ID])
- **[OTHER_TYPE]**: [DEFINITION] ([TABLE]: [ID])

**Relationships:**
- [ENTITY_1] → [ENTITY_2]: [RELATIONSHIP_TYPE] (join on [JOIN_KEY])

---

## Business Terminology

| Term | Definition | Notes |
|------|------------|-------|
| [TERM_1] | [DEFINITION] | [CONTEXT/GOTCHA] |
| [TERM_2] | [DEFINITION] | [CONTEXT/GOTCHA] |
| [ACRONYM] | [FULL_NAME] - [EXPLANATION] | |

---

## Standard Filters

Always apply these filters unless explicitly told otherwise:

-- Exclude test/internal data WHERE [TEST_FLAG_COLUMN] = FALSE AND [INTERNAL_FLAG_COLUMN] = FALSE

-- Exclude invalid/fraud AND [STATUS_COLUMN] != '[EXCLUDED_STATUS]'

-- [OTHER STANDARD EXCLUSIONS]


**When to override:**
- [SCENARIO_1]: Include [NORMALLY_EXCLUDED] when [CONDITION]

---

## Key Metrics

### [METRIC_1_NAME]
- **Definition**: [PLAIN_ENGLISH_EXPLANATION]
- **Formula**: `[EXACT_CALCULATION]`
- **Source**: `[TABLE_NAME].[COLUMN_NAME]`
- **Time grain**: [DAILY/WEEKLY/MONTHLY]
- **Caveats**: [EDGE_CASES_OR_GOTCHAS]

### [METRIC_2_NAME]
[REPEAT FORMAT]

---

## Data Freshness

| Table | Update Frequency | Typical Lag |
|-------|------------------|-------------|
| [TABLE_1] | [FREQUENCY] | [LAG] |
| [TABLE_2] | [FREQUENCY] | [LAG] |

To check data freshness:

SELECT MAX([DATE_COLUMN]) as latest_data FROM [TABLE]


---

## Knowledge Base Navigation

Use these reference files for detailed table documentation:

| Domain | Reference File | Use For |
|--------|----------------|---------|
| [DOMAIN_1] | `references/[domain1].md` | [BRIEF_DESCRIPTION] |
| [DOMAIN_2] | `references/[domain2].md` | [BRIEF_DESCRIPTION] |
| Entities | `references/entities.md` | Entity definitions and relationships |
| Metrics | `references/metrics.md` | KPI calculations and formulas |

---

## Common Query Patterns

### [PATTERN_1_NAME]

[SAMPLE_QUERY]


### [PATTERN_2_NAME]

[SAMPLE_QUERY]


---

## Troubleshooting

### Common Mistakes
- **[MISTAKE_1]**: [EXPLANATION] → [CORRECT_APPROACH]
- **[MISTAKE_2]**: [EXPLANATION] → [CORRECT_APPROACH]

### Access Issues
- If you encounter permission errors on `[TABLE]`: [WORKAROUND]
- For PII-restricted columns: [ALTERNATIVE_APPROACH]

### Performance Tips
- Filter by `[PARTITION_COLUMN]` first to reduce data scanned
- For large tables, use `LIMIT` during exploration
- Prefer `[AGGREGATED_TABLE]` over `[RAW_TABLE]` when possible

Raw lines 1 to 139

140## Customization Notes

141

142When generating a skill:

143

1441. **Fill all placeholders** - Don't leave any `[PLACEHOLDER]` text

1452. **Remove unused sections** - If they don't have dashboards, remove that section

1463. **Add specificity** - Generic advice is less useful than specific column names and values

1474. **Include real examples** - Sample queries should use actual table/column names

1485. **Keep it scannable** - Use tables and code blocks liberally

Customization Notes

When generating a skill:

Fill all placeholders - Don't leave any [PLACEHOLDER] text
Remove unused sections - If they don't have dashboards, remove that section
Add specificity - Generic advice is less useful than specific column names and values
Include real examples - Sample queries should use actual table/column names
Keep it scannable - Use tables and code blocks liberally

Raw lines 140 to 148