kanari audit

One-shot health check with exit codes for CI/CD integration

Overview

kanari audit connects to your Redis and Celery instances, collects metrics, runs the findings engine, and prints a formatted report. It then exits with a status code that reflects the severity of what it found.

kanari audit

---

Flags

Flag	Default	Description
--config, -c	—	Path to a YAML config file
--json	off	Output machine-readable JSON instead of TUI
--md	off	Output a Markdown report
--no-config-checks	off	Skip Redis/Celery configuration analysis (runs by default)
--deep, -d	—	Deprecated no-op — configuration analysis now runs by default
--timeout	3.0	Max seconds to wait for connections and metrics
--no-color	off	Disable ANSI colors (useful in CI logs)

---

Exit codes

Code	Meaning
0	Healthy — no findings
1	Warnings — at least one MEDIUM or LOW finding
2	Critical — at least one HIGH or CRITICAL finding

---

Output modes

TUI (default)

Rich-formatted report with color-coded status indicators. Best for humans running audits interactively.

kanari audit

JSON

Machine-readable output. Use this in CI scripts, log aggregation, or when piping to other tools.

kanari audit --json

{
  "timestamp": "2026-05-31T14:22:01Z",
  "system_status": "DEGRADED",
  "exit_code": 2,
  "top_findings": [
    {
      "id": "WORKER_OFFLINE",
      "severity": "CRITICAL",
      "title": "Worker offline: beat-wrk",
      "evidence_summary": "worker=beat-wrk, is_alive=False"
    },
    {
      "id": "QUEUE_SLA_BREACH_EMAILS",
      "severity": "HIGH",
      "title": "SLA breach on queue 'emails'",
      "evidence_summary": "queue=emails, latency_sec=125.4"
    }
  ],
  "metrics": {
    "total_pending": 862,
    "total_active": 7,
    "saturation_pct": 43.8,
    "max_latency_sec": 125.4,
    "redis_connected": true,
    "celery_connected": true
  }
}

Markdown

Structured Markdown output. Useful for posting to Slack, GitHub issues, or incident reports.

kanari audit --md

---

Config file auto-discovery

--config is optional. When omitted, kanari audit looks for kanari.yaml in the current directory and loads it automatically:

kanari audit
# 💡 Using kanari.yaml

If kanari.yaml is not present, it falls back to localhost defaults and suggests creating one:

kanari audit
# 💡 No kanari.yaml found — connecting to localhost defaults.
#    Run 'kanari init' to generate one.

To use a config file in a different location, pass it explicitly:

kanari audit --config /etc/kanari/prod.yaml

Run kanari init to generate kanari.yaml, or see Configuration for the full reference.

---

Configuration analysis

Every audit includes a Configuration Analysis section that checks your Redis and Celery settings for common production risks — no flag needed. Use --no-config-checks to skip it. See Configuration Analysis for details.

---

Examples

# Quick check — is anything broken right now?
kanari audit

# CI gate — fail pipeline if critical issues found
kanari audit --json && echo "Celery healthy"

# Debug an incident, saving the full report
kanari audit --no-color 2>&1 | tee incident-$(date +%Y%m%d).log

# Check a specific environment
REDIS_URL=redis://staging-redis:6379/0 kanari audit