Open source to start. Credits for what you actually use.
Four tiers. OSS for every platform team. Cloud when you want the dashboard. Growth when you want the ceiling and priority. Custom when procurement sends a questionnaire.
OSS
Open sourceMIT-licensed SDK, Gateway, and MCP Server. Local policy evaluation. No hosted dashboard.
- bookbag-qa-sdk (Python + Node, zero-dep)
- bookbag-gateway-sdk (MIT, Docker)
- bookbag-mcp-server (MIT, Claude Desktop / Cursor)
- Local policy evaluator (JSON-match rules)
- Default policy bundle (destructive shell, SQL drop, prod writes)
- Hosted dashboard (Observe / Guardrails / Evaluation UIs)
- Framework evidence exports
- Staged AI auditor
Cloud
CloudHosted control plane + credit slider. For teams shipping AI to production today.
- Everything in OSS
- Hosted traces + live activity feed (SSE)
- Guardrails rule authoring UI
- Observe dashboards (inventory, runs, activity)
- Framework mapping (EU AI Act, NIST AI RMF, ISO 42001, SOC 2)
- Staged AI auditor — standard tier
- Evidence exports (signed manifest)
- 3 seats · email support
Growth
GrowthHigher credit ceiling, priority queue, SSO on request. For teams scaling AI across the org.
- Everything in Cloud
- Slider ceiling up to 60,000 credits/mo
- Priority queue on the AI auditor
- Staged AI auditor — fast / standard / deep
- Eval harness + GitHub Action CI integration
- Early access to bundled red-team suites (ships 2026 Q3)
- HIPAA Trust Services Criteria mapping
- SSO on request · 5 seats · priority support
Custom
ScaleOn-prem deployment, custom framework controls, 99.95% SLA. For regulated enterprises.
- Everything in Growth
- Unlimited credits · custom retention
- On-prem (Docker / Helm) or dedicated VPC
- Custom framework controls + policy packs
- Dedicated compliance representative
- 99.95% uptime SLA · 24/7 support
- BAA / DPA / custom security reviews
- SSO on request · unlimited seats
Pick your credits
Platform subscription is flat. AI usage is metered in credits. Slide to see what you'd pay — Cloud fits 100–10K, Growth goes up to 60K.
Pick your credits
AI-auditor usage is priced by credit. Guardrails rule eval is unmetered — credits fire only when an AI model runs.
Credits renew monthly. Platform subscription (Cloud) billed separately.
Rates shown are a preview. Your account page in the app shows live rates and is the source of truth.
Auto-mapped evidence, not another PDF audit
What every paid tier ships with
These are table stakes on Cloud and up — OSS gives you the SDK layer.
Live activity + causal run trace
Runtime block / hold / redact / flag
Taxonomy-driven output scoring
Framework controls + audit trail
Pricing FAQs
Frequently Asked Questions
Start free. Upgrade when the value is obvious.
Join the teams shipping safer AI with real-time evaluation, audit trails, and continuous improvement.