Showcase agents / Incident Investigator
SRE / on-call
Incident Investigator
What problem? Incident bots produce long tool chains with no trace depth visibility — wrong RCA hypotheses look identical to correct ones in logs.
SRE reference: alert → metrics/logs/deploys → RCA hypothesis → draft PR. Long chains with Claude-primary reasoning.
What Inferix observes
Every run dual-writes to LensAI and TraceForge. Switch to this tenant in the console to see live KPIs.
- TraceForge — deep tool waterfall and chain depth
- LensAI — token cost per investigation
- DriftWatch — RCA quality vs golden set
- FineForge — promote better retrieval prompts
Tools & models
Real SQLite backends, not mocked HTTP stubs.
- pager.get_alert
- metrics.query
- logs.search
- git.recent_deploys
- k8s.get_events
- github.open_draft_pr
Model routing
- Claude (long context, reasoning)
- limited cheap path
Try it
Start the stack, then run the golden scenario. Traffic appears in the console within seconds.
1. Start stack + agents
cd inferix && ./run-inferix.sh --full --agents
2. Golden scenario
curl -s -X POST http://localhost:9104/scenarios/latency_regression | jq
Expect HTTP 200 with JSON status and tool steps. Traces appear under this tenant in ~10s.
3. All agents at once
cd inferix && ./scripts/generate-agent-traffic.sh
Expected: JSON with status and tool steps. Then open the tenant console and filter traces by tenant-incident-ops. In-browser runner stays deferred — curl is the verified path.
Console: /console/overview?tenant_id=tenant-incident-ops · API traces: GET /v1/traces?tenant_id=tenant-incident-ops
See all seven agents in the console
Multi-tenant super-dashboard. Filter by tenant, drill into traces and costs.