| name | govern-artifact |
| description | Attach governance metadata (provenance, lineage, confidence, evidence) to any artifact. Use this atom when document-intelligence, output-validator, or any skill needs to stamp an artifact with its origin, processing chain, and confidence level. Do NOT use for quality gating — that is atom-quality-check. Do NOT use for content generation. |
govern-artifact
Stamps any artifact with governance metadata: provenance (where it came from), lineage (how it was processed), confidence (how reliable), and evidence (what supports it). Used by document-intelligence after parsing and reusable by any skill producing governed artifacts.
Input
{
"artifact": {"type": "lesson_plan", "content": "..."},
"source_lineage": {
"origin": "document_intelligence",
"source_file": "curriculum.pdf",
"extracted_at": "2026-06-27T10:00:00Z",
"parser_used": "PyMuPDF"
},
"confidence": {"overall": 0.95, "text_recovery": 0.98}
}
Output
{
"tool": "govern-artifact",
"governed_artifact": {
"type": "lesson_plan",
"content": "...",
"_provenance": {"origin": "document_intelligence", "source_file": "curriculum.pdf"},
"_lineage": {"extracted_at": "2026-06-27T10:00:00Z", "parser_used": "PyMuPDF"},
"_confidence": {"overall": 0.95, "text_recovery": 0.98}
},
"human_review_required": true
}
Do NOT use this atom for
- Quality gating or pass/fail validation (use atom-quality-check)
- Content generation or modification (this only adds metadata)
- Security scanning
Pipeline note
Follows references/method.md at the Governance step (metadata attachment). Output conforms to references/metadata-schema.md. human_review_required: true — provenance and confidence metadata should be verified.
