Daemon

The Fuze daemon is an optional background process that adds cross-run pattern detection, org-wide budget enforcement, kill switches, and persistent audit storage.

Starting the daemon

npx fuze-ai daemon

The daemon listens on a Unix Domain Socket (or Windows named pipe) and exposes an HTTP API for the dashboard and external integrations.

What the daemon adds

CapabilitySDK onlySDK + Daemon
Per-run budgetYesYes
Org-wide daily budgetNoYes
Per-agent daily budgetNoYes
Loop detectionPer-runPer-run + cross-run patterns
Trace storageJSONL fileSQLite with hash-chained audit log
Kill switchNoYes — via API or dashboard
Compensation/rollbackIn-process onlyPersistent + API-triggered
AlertsNoWebhooks + dashboard WebSocket
Pattern analysisNoRepeated failures, cost spikes, reliability drops

Architecture

┌─────────────┐     UDS/pipe      ┌──────────────┐     HTTP     ┌───────────┐
│  Your Agent │  ──────────────── │  Fuze Daemon │ ──────────── │ Dashboard │
│  (fuze-ai)  │                   │              │              │  (React)  │
└─────────────┘                   │ ┌──────────┐ │              └───────────┘
                                  │ │  SQLite  │ │
                                  │ │  audit   │ │
                                  │ └──────────┘ │
                                  └──────────────┘

The SDK communicates with the daemon over a Unix Domain Socket (Linux/macOS) or named pipe (Windows). The daemon stores all audit data in SQLite with SHA-256 hash chains for tamper detection.

Configuration

[daemon]
socket_path = "/tmp/fuze.sock"       # Unix socket path
api_port = 7700                       # HTTP API port
storage_path = "~/.fuze/traces.db"    # SQLite database
retention_days = 180                  # Min 180 for EU AI Act Art. 19

[daemon.budget]
org_daily_budget = 100.00             # Org-wide daily ceiling (USD)
per_agent_daily_budget = 20.00        # Per-agent daily ceiling (USD)
alert_threshold = 0.80                # Alert at 80% of ceiling

[daemon.alerts]
dedup_window_ms = 60000               # Suppress duplicate alerts within window
webhook_urls = ["https://hooks.slack.com/..."]

On Windows, use a named pipe:

[daemon]
socket_path = "\\\\.\\pipe\\fuze.sock"

API endpoints

MethodPathDescription
GET/api/healthDaemon liveness check
GET/api/runsPaginated run list with filters
GET/api/runs/:idSingle run with steps and events
POST/api/runs/:id/killKill an active run
GET/api/runs/:id/compensationCompensation records for a run
POST/api/runs/:id/rollbackTrigger manual rollback
GET/api/budgetOrg and per-agent spend
GET/api/agents/:id/healthAgent reliability stats
GET/api/compliance/report/:idEU AI Act incident report
WS/wsLive alerts stream

Audit integrity

Every record in the SQLite database is hash-chained using SHA-256. The daemon maintains four independent chains:

  1. Runs — immutable fields hashed at insertion
  2. Steps — every tool call recorded and chained
  3. Guard events — loop detections, budget blocks, kills
  4. Compensation records — rollback actions and outcomes

Verify chain integrity:

# Via API
curl http://localhost:7700/api/compliance/report/RUN_ID

The verifyHashChain() method checks all four chains and reports the first broken record if tampering is detected.

Data retention

The daemon automatically purges data older than retention_days. Purging cascades across all related tables:

  • Runs
  • Steps
  • Guard events
  • Compensation records
  • Idempotency keys

This ensures no orphaned records remain after purge.

Resource management

The daemon caps in-memory ended runs at 1,000 entries (FIFO eviction) to prevent unbounded memory growth during long-running deployments. Active runs are always retained.

Side-Effect TrackingDashboard