tg-show-explain-trace

Shows the full explainability trace for a query session.

Synopsis 

tg-show-explain-trace QUESTION_ID [options]

Description

Displays the complete reasoning trace for a GraphRAG, DocRAG, or Agent session. Auto-detects the trace type.

  • GraphRAG: Question -> Grounding -> Exploration -> Focus -> Synthesis
  • DocRAG: Question -> Exploration -> Synthesis
  • Agent: Session -> Iterations (thought/action/observation) -> Conclusion

With --show-provenance, traces selected edges back to their source documents via reification.

Options

Required Arguments

Option Description
QUESTION_ID Question or session URI to show trace for

Optional Arguments

Option Default Description
-u, --api-url URL $TRUSTGRAPH_URL or http://localhost:8088/ API URL
-t, --token TOKEN $TRUSTGRAPH_TOKEN Authentication token
-f, --flow-id ID default Flow ID
-U, --user USER trustgraph User identifier
-C, --collection COLL default Collection identifier
--max-answer N 500 Max characters for answer display
--show-provenance false Trace edges back to source documents
--format FORMAT text Output format: text, json

Examples 

tg-show-explain-trace "urn:trustgraph:question:abc123"

tg-show-explain-trace --show-provenance "urn:trustgraph:question:abc123"

tg-show-explain-trace --format json "urn:trustgraph:agent:abc123"

Environment Variables

  • TRUSTGRAPH_URL: Default API URL
  • TRUSTGRAPH_TOKEN: Default authentication token