Cova CLI

Monitor your monitoring from the terminal. Scan repos, connect tools, run analyses, and check PRs for monitoring gaps - all from the command line or CI/CD pipelines.

Installation

pip install cova-cliCopy

Requires Python 3.9+. Works on macOS, Linux, and Windows.

Verify the installation:

cova --version
# cova-cli 0.1.0

Authentication

1. Generate a token

Go to Settings > General > CLI & API Access in the Cova web app and click Generate Token. Copy the token - it's only shown once.

2. Log in

cova login
# Enter your API token: ****
# Logged in as you@email.com (pro)Copy

Or pass the token directly (useful for scripts):

cova login --token cova_your_token_here

3. Log out

cova logout

Removes the saved token from ~/.cova/config.

Commands

cova status

Show your account info, plan, and connected tools.

cova status

  Account:  you@email.com
  Plan:     pro
  Tools:    Datadog, PagerDuty, SentryCopy

cova scan

Scan a local repository for monitoring recommendations. Analyzes your source code to identify endpoints, databases, services, and suggests what to monitor.

# Scan current directory
cova scan .

# Scan a specific repo
cova scan ~/projects/my-api

# JSON output for scripting
cova scan . --jsonCopy

The scan archives your source files (skipping node_modules, .git, venv, binaries, etc.) and uploads them for AI-powered analysis.

cova connect

Connect a monitoring tool. Prompts for the required credentials interactively.

cova connect datadog
#   API Key: ****
#   Application Key: ****
#   Datadog Site [datadoghq.com]: us5.datadoghq.com
# Connected to Datadog. 47 monitors found.Copy

Supported tools:

cova analyze

Run a monitor analysis across all connected tools. Returns a health score, coverage breakdown, and findings.

cova analyze

# ╭──────── Health Score ────────╮
# │     72/100  Needs Work       │
# ╰──────────────────────────────╯
# 
# Coverage
#   Uptime                ████████████████░░░░ 78%
#   Error Tracking        ██████████░░░░░░░░░░ 50%
#   Alerting              ████████████████████ 100%Copy

cova pr-check CI/CD

Check a git diff for monitoring gaps. Runs locally using git diff - no GitHub or GitLab connection required. Returns exit code 0 if clean, 1 if gaps are found - perfect for CI gates.

# Check latest commit
cova pr-check

# Check against main branch
cova pr-check --diff main..HEAD

# Check specific range
cova pr-check --diff HEAD~3

# Pipe a diff from stdin
git diff main | cova pr-check --from-stdinCopy

Exit codes:

Global Options

These flags work with every command:

Examples:

# JSON output, pipe to jq
cova analyze --json | jq '.health_score'

# Quiet mode - only exit code matters
cova pr-check -q && echo "Clean" || echo "Gaps found"

# Per-command help
cova scan --help

CI/CD Integration

GitHub Actions

# .github/workflows/cova.yml
name: Monitoring Check
on: [pull_request]

jobs:
  cova:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
        with:
          fetch-depth: 0

      - name: Install Cova CLI
        run: pip install cova-cli

      - name: Check for monitoring gaps
        run: cova pr-check --diff origin/main..HEAD --json
        env:
          COVA_TOKEN: ${{ secrets.COVA_TOKEN }}Copy

GitLab CI

# .gitlab-ci.yml
cova-check:
  image: python:3.11-slim
  script:
    - pip install cova-cli
    - cova pr-check --diff origin/main..HEAD --json
  variables:
    COVA_TOKEN: $COVA_TOKENCopy

Pre-commit Hook

# Run before every push
echo 'cova pr-check -q || echo "Warning: monitoring gaps detected"' >> .git/hooks/pre-push
chmod +x .git/hooks/pre-push

CLI vs Web App

The CLI gives you access to the same core Cova features as the web app - same backend, same analysis engine. The main differences are in how you interact with the results.

What works the same

• Tool connections are shared. Connect Datadog via CLI and it shows up in the web app (and vice versa). Same account, same credentials.
• Scan results are shared. Run cova analyze from the CLI and the results appear in the web dashboard too.
• Same analysis engine. CLI and web use the exact same backend API and AI models. No difference in accuracy or coverage.

What's different

• PR Guard comments. The web-based PR Guard posts detailed fix recommendations directly on your PRs as comments. The CLI pr-check only outputs to your terminal/CI logs and returns an exit code. For inline PR feedback, use the web PR Guard with GitHub or GitLab connected.
• Interactive features. Config generation, deploy-to-tool, incident investigation, and chat require the web UI. The CLI is optimized for scanning, checking, and CI/CD automation.
• No GitHub/GitLab required. The CLI works with local git repos directly. You don't need to connect a SCM provider to use cova scan or cova pr-check.

Configuration

The CLI stores its config at ~/.cova/config (permissions 600):

{
  "token": "cova_...",
  "api_url": "https://nocta-backend.onrender.com"
}

Environment Variables

Environment variables always take priority over the config file.

Quick Reference

# Setup
pip install cova-cli           # install
cova login                     # authenticate (interactive)
cova status                    # check account + tools

# Connect tools
cova connect datadog           # connect Datadog
cova connect pagerduty         # connect PagerDuty
cova connect sentry            # connect Sentry
cova connect grafana           # connect Grafana
cova connect newrelic          # connect New Relic
cova connect sumologic         # connect Sumo Logic
cova connect splunk            # connect Splunk

# Scan + Analyze
cova scan .                    # scan repo for monitoring recs
cova scan ~/my-api             # scan a specific path
cova analyze                   # run monitor analysis
cova pr-check                  # check latest commit for gaps
cova pr-check --diff main..HEAD  # check branch diff

# Output modes
cova analyze --json            # machine-readable JSON
cova pr-check -q               # quiet (exit code only)
cova scan . --json | jq        # pipe to jq

# CI/CD (no login needed)
COVA_TOKEN=xxx cova pr-check --json  # env var auth
cova logout                    # clear saved tokenCopy

Troubleshooting