Dashboard¶
AgentSniff includes a web dashboard for real-time scan monitoring, history browsing, and settings management.
Starting the Dashboard¶
Loopback use is credential-free. When AGENTSNIFF_API_TOKEN is configured, the dashboard is protected with HTTP Basic authentication: enter agentsniff as the username and the token as the password. AgentSniff refuses non-loopback binding without a token; use TLS for any remote listener.
Features¶
Live Scanning¶
- Select target network and detectors from the UI
- Real-time agent detection via Server-Sent Events (SSE)
- Presence and trusted-identity counters update as assets are found
- Stop button cancels the scan immediately
Scan Results¶
- Service assets displayed with framework, host, port, presence, identity assurance, risk, and change state
- Expandable signal details for each detected agent
- Presence confirmation is visually separate from cryptographic identity verification
Scan History¶
- Previous scans stored in SQLite or PostgreSQL
- Browse past results with timestamps, asset counts, and new/changed/removed totals
- Stable service IDs and first-seen timestamps carry across scans of the same target
- Cancelled and incomplete scans are preserved
Settings¶
Click the gear icon to configure:
- Changes only — Notify only for new, changed, or removed assets (enabled by default)
- Alert thresholds — Minimum assets and presence confidence needed to trigger alerts
- Webhook URL — HTTP POST endpoint for alert notifications
- SMTP settings — Email server configuration for alert emails
- Test Alert — Send a test notification to verify configuration
Export & Database Management¶
The Export dropdown writes the latest scan in JSON, CSV, or SARIF 2.1.0 format.
On v2, the dashboard also exposes database administration controls backed by GET /api/db/backup and POST /api/db/reset:
- Backup Database — downloads a SQLite snapshot or PostgreSQL JSON scan export
- Reset Database — wipes all stored scans (confirmation required)