| name | federation-traces |
| description | Ingest GA trace artifacts, analyze with ix stats, feed results to TARS pattern promotion |
Federation Trace Pipeline
Cross-repo trace pipeline: GA exports, ix analyzes, TARS promotes patterns.
When to Use
When trace data from GA needs statistical analysis before being fed into TARS's pattern promotion pipeline.
Pipeline
Step 1: Prepare traces from GA
Tool: ix_tars_bridge with action=prepare_traces
- Loads traces from
~/.ga/traces/
- Computes stats (latency, event distribution, anomalies)
- Returns payload formatted for TARS ingestion
Step 2: Analyze traces
Tool: ix_trace_ingest
- Loads trace directory, computes detailed stats
- Mean/median/p95 latency per event type
- Success/failure rates
- Event frequency distribution
Step 3: Ingest into TARS
Tool: TARS ingest_ga_traces
- Input:
{"Count": N, "MinOccurrences": 3}
- Runs pattern discovery on trace artifacts
- Promotes recurring patterns up the staircase
Step 4: View promotion results
Tool: TARS promotion_index
- Returns ranked patterns sorted by level, score, weight
Step 5: Export insights
Tool: TARS export_insights
- Writes
~/.tars/insights/latest.json with pattern scores, gaps, recommendations
Trace Format (GA)
{
"trace_id": "...",
"timestamp": "...",
"events": [{ "type": "...", "duration_ms": 0, "metadata": {} }],
"outcome": "success|failure"
}
Analysis Output (ix)
- Mean/median/p95 latency per event type
- Event frequency distribution
- Success/failure correlation analysis