Find root cause before the retro does.
TRCAA is a structured, AI-backed desktop tool for incident triage, 5-Whys root cause analysis, and blameless post-mortems — fully offline via Ollama, or connected to the cloud provider of your choice.
The workflow
Six steps from alert to after-action.
- 01
New Issue
Select domain, enter title and severity
- 02
Log Upload
Drag-and-drop log files, review PII redactions
- 03
Triage
5-Whys AI conversation, auto-tracked why levels 1–5
- 04
Resolution
Review and confirm each root cause and action
- 05
RCA
Auto-generated RCA document, export as MD or PDF
- 06
Post-Mortem
Blameless post-mortem document with action items
PII sanitization
Nothing leaves the machine unredacted.
Every log is scanned by a Rust-side, multi-pattern PII detector before a single byte reaches an AI provider. Spans are reviewed and approved in the UI, and every external send is logged with a SHA-256 hash for audit.
Shell execution
Agentic shell access, with a safety governor built in.
kubectl, Proxmox tools, and general diagnostics run under a three-tier classification system. Nothing destructive ever runs without a human in the loop — and some things never run at all.
- Tier 1 — Auto-execute
kubectl get/describe/logs, cat, grep, ls
- Tier 2 — User approval
kubectl apply/delete/scale, ssh, systemctl restart
- Tier 3 — Always denied
rm -rf, shutdown, mkfs
Domain coverage
Sixteen expert system prompts, ready on day one.
Runs offline. Scales to the cloud.
Start free with the MIT core. Add cloud AI, team workflows, or enterprise compliance when you need them.