Artisan commands

Screening & output

# Screen a prompt (exits non-zero when blocked); reads STDIN if no argument
php artisan ai-guardrails:screen "please ignore all previous instructions"

# Sanitize + redact a text blob
php artisan ai-guardrails:sanitize "<script>steal()</script> ![x](http://evil/leak)"

# List recent injection-audit attempts (blocked and allowed)
php artisan ai-guardrails:audit --limit=50

Retention (GDPR)

# Apply the configured retention strategy to the audit table AND the HITL sidecar (actor-audited)
php artisan ai-guardrails:purge --strategy=anonymize --days=365 --actor="ops:nightly"

# Preview what would be affected, change nothing
php artisan ai-guardrails:purge --dry-run

The command sweeps every table whose store is set to database in a single actor-audited run (ai_guardrails_injection_audit when audit.store=database, ai_guardrails_hitl_requests when hitl_requests.store=database). Anonymize on the sidecar redacts arguments to {} and nulls principal_id while keeping the approval trail (tool, run_id, occurred_at).

See audit hygiene & retention for the full semantics.

HITL setup

# Run laravel-flow's migrations (flow_runs / flow_approvals) scoped from vendor — idempotent
php artisan ai-guardrails:hitl-install

# Diagnose the HITL setup; non-zero exit until it can actually gate a call
php artisan ai-guardrails:hitl-status

MCP (when enabled)

# Start the local (stdio) MCP server exposing the guardrail tools
php artisan mcp:start ai-guardrails