Skip to main content

Quick start

Get from zero to your first AI-SDLC scan in about five minutes. No account or network needed to scan — your code stays on your machine.

For the full DX contract (IDE, CI, tokens, Monitor Mode), see Developer experience.

1. Install the CLI​

Works today (design partners): use the checksummed binary your Unveilr contact hands you, or — with private-repo access — build from source:

# Hand-delivered binary
shasum -a 256 -c unveilr.sha256
chmod +x unveilr && sudo mv unveilr /usr/local/bin/
unveilr --help

# Or, operators / partners with monorepo access:
make install && apps/api/.venv/bin/unveilr --help
CDN installer — after GA

curl -fsSL https://get.unveilr.ai/install.sh | sh is provisioned as part of the public cut. Until DNS/CDN is live it will not resolve — do not lead with it. Full paths: Installation.

2. Scan a repository​

cd your-repo
unveilr scan
UNVEILR — AI-SDLC scan of .

AI-BOM (10 assets)
  ai_tool 1 · dependency 6 · mcp_server 1 · model_ref 1 · prompt_file 1

Findings (13):  critical 1 · high 10 · medium 2
  [CRITICAL] secret_leak    handler.py:5      AWS access key id
  [HIGH]     insecure_code  app.js:5          Use of eval()
  [HIGH]     iac_exposure   main.tf:18        Security group open to the world (0.0.0.0/0)
  [HIGH]     typosquat      requirements.txt  'reqests' looks like a typosquat of 'requests'

That single command did Discover (the AI-BOM) and Guard (the findings).

Screenshot to capture — CLI output

Where: Terminal running unveilr scan on a controlled test repository. Filename: cli-scan-terminal.png — see Screenshot guide.
Story it tells: “Value in one command, offline.”

3. Add it to CI (safely)​

Monitor Mode never fails a build, so you can add it everywhere today:

unveilr scan                       # observe — always exits 0
unveilr scan --mode enforce --fail-on high   # gate — exits 1 on high+ findings

Or one GitHub Action block:

- uses: unveilr/scan-action@v1
  with:
    mode: observe
    fail-on: high

4. See it with your team (optional)​

export UNVEILR_API=https://guard.unveilr.ai   # or your operator’s origin
unveilr login --token "$UNVEILR_TOKEN" --api "$UNVEILR_API"
# CI: unveilr login --token "$UNVEILR_TOKEN" --api "$UNVEILR_API"
unveilr scan --upload

Then open Overview and AI Inventory at https://guard.unveilr.ai — or follow Get started (/onboarding). See Production APIs and Persona playbooks for every role.

5. Govern agents when you're ready​

Register agents, approve scopes, and point MCP clients at the Gateway — see Govern. No need to block on this for your first scan.

Where to next​