{
  "schema_version": "1.1",
  "id": "archive:https://www.youtube.com/watch?v=H7puB0RwJMM",
  "slug": "citation-needed-provenance-for-llm-built-knowledge-graphs-daniel-chalef-1iob5t8",
  "url": "https://feed7.dev/p/citation-needed-provenance-for-llm-built-knowledge-graphs-daniel-chalef-1iob5t8",
  "title": "Citation Needed: Provenance for LLM-Built Knowledge Graphs — Daniel Chalef, Zep AI",
  "why_included": "LLM synthesis breaks simple source pointers when facts merge, change, or outlive one input. Model provenance inside the context graph so agents can filter trust, debug derivation, and apply deletion policies.",
  "summary": "Graphiti stores source episodes and links derived entities and facts back to them. When entities merge or new evidence contradicts a fact, those links and temporal changes preserve **fact-level lineage** instead of relying on a later log.",
  "practical_implication": "Build provenance into an agent-memory data structure at ingestion. Propagate source metadata, filter retrieval by trust policy, and use remaining source links to decide whether a derived fact survives **source deletion**.",
  "agent_context": "Graphiti stores source episodes and links derived entities and facts back to them. When entities merge or new evidence contradicts a fact, those links and temporal changes preserve **fact-level lineage** instead of relying on a later log.\n\nBuild provenance into an agent-memory data structure at ingestion. Propagate source metadata, filter retrieval by trust policy, and use remaining source links to decide whether a derived fact survives **source deletion**.\n\nVerification policy is contextual: some decisions require any verified parent, while others require every parent to be verified. Graph construction is also **expensive in cost and latency**, and one discussed many-edge case uses a separate structure rather than keeping all provenance in the graph.",
  "source": {
    "name": "AI Engineer",
    "url": "https://www.youtube.com/watch?v=H7puB0RwJMM",
    "published_at": "2026-07-23T04:00:02.000Z"
  },
  "source_class": "video",
  "content_type": "Video",
  "layer": "context",
  "domains": [
    "data",
    "security"
  ],
  "topics": [
    "agent-memory",
    "retrieval",
    "context-engineering"
  ],
  "verification": {
    "status": "source_linked",
    "label": "Source Linked",
    "method": "source_feed",
    "verified_at": null
  },
  "uncertainty": [
    "Verification policy is contextual: some decisions require any verified parent, while others require every parent to be verified. Graph construction is also **expensive in cost and latency**, and one discussed many-edge case uses a separate structure rather than keeping all provenance in the graph."
  ],
  "connected_context": null,
  "lifecycle": "Current",
  "published_at": "2026-07-23T04:00:02.000Z",
  "modified_at": "2026-07-23T04:00:02.000Z",
  "supersedes": [],
  "expires_at": null,
  "formats": {
    "html": "https://feed7.dev/p/citation-needed-provenance-for-llm-built-knowledge-graphs-daniel-chalef-1iob5t8",
    "json": "https://feed7.dev/p/citation-needed-provenance-for-llm-built-knowledge-graphs-daniel-chalef-1iob5t8.json",
    "markdown": "https://feed7.dev/p/citation-needed-provenance-for-llm-built-knowledge-graphs-daniel-chalef-1iob5t8.md"
  }
}