.webp&w=64&q=75&dpl=dpl_CpQbZ7uW7SJY3tWJBtJsf3EBnJyp){kind=small}
SigillumOne pricing engine
Do not fork pricing math per persona. Quotes should remain the single source of truth, with different packaging and admin language for maintainers, indie builders, and autonomous agents.
PRICING
Commercial admin
Pricing
Sigillum should keep one shared quote engine underneath every persona. This page now presents that model in a maintainable admin language: what gets priced, how teams should think about budgets, and how pricing should be explained without exposing x402 internals.
Maintainer surface
What this page owns today
These cards translate the live Sigillum product state into a maintainer/admin runbook without exposing raw payloads or editing the payment path.
Pay-as-you-go first
The existing x402 flow already supports per-action charging. That is enough to productize before adding subscriptions or credit pools.
Budget UI still missing
Teams need a clearer story for max spend per action and rolling budget caps. Those controls belong in the admin surfaces even before formal billing infrastructure exists.
Primary runbook
Shared pricing model to operationalize
This is the smallest serious pricing posture that stays honest to the existing product and keeps the quote service as the canonical source of truth.
Price from inspected work, not persona
Every quote should still derive from the work Sigillum is being asked to inspect: action type, inspected units, and any depth multiplier you intentionally expose.
- Keep the base quote attached to the action envelope and unit estimation.
- Use action-type multipliers only when the model truly needs them.
- Show the resulting paid amount back in receipts and the dashboard exactly as charged.
Package the same math differently
The commercial framing should feel native to the persona without creating separate hidden pricing systems.
- Maintainers think in per-PR cost and repo budgets.
- Indie builders think in one-shot inspect cost and simple personal caps.
- Autonomous agents think in rolling team budgets and automated spend guardrails.
Budget before subscriptions
The next production-ready improvement is clear budget and spend policy, not a large billing-platform redesign.
- Max spend per action prevents surprise inspect costs.
- Daily and monthly caps create predictable team behavior.
- Budget-triggered stops should be auditable in the same proof language as risk-triggered stops.
Reference snippets
Copy-safe examples
Keep these snippets aligned with the current hosted Sigillum flow and existing workflow inputs. They are admin guidance only, not mock runtime data.
Quote fields to preserve in admin messaging
quote_id=quo_... action_id=act_... amount=<real quote> USDC expires_at=<timestamp> inspected_units=<persisted unit snapshot>
Admin checks
What pricing should already communicate
These are the truths the admin surface should make obvious to operators before expanding into subscriptions or credits.
TruthQuotes are real, not dummy placeholders. They are derived from the inspected action and should be treated as the source of truth for amount before payment.
ClarityReceipts and dashboard cards should echo the actual paid amount and payment reference so teams can reconcile inspection spend after the fact.
BudgetThe missing production piece is operator-facing budget policy, especially max spend per action and rolling caps by repo, user, or agent.
FutureCredits or subscriptions can come later, but they should wrap the same quote engine rather than replace it with persona-specific hidden pricing logic.
Admin metrics
What maintainers should watch
These are the operational proof points the current GitHub, policy, and pricing surfaces should make easy to audit before expanding scope.
Quote truth
Action + units
The price should be explained as a function of action type and inspected units, not as a magic fixed fee disconnected from the quote response.
Maintainer packaging
Per PR
GitHub maintainers mostly care about predictable PR review spend and repo-level caps rather than wallet mechanics.
Builder packaging
Per inspect
Solo builders need a simple hosted CLI price story with as little gateway jargon as possible.
Agent packaging
Per action stream
Automated systems care about policy and budget envelopes across many actions rather than one-off receipt totals.
Pricing packaging
Persona-facing commercial framing
One quote engine still prices the work. These cards show how that same logic can be presented to different operators without introducing separate backend math.
GitHub maintainer
Pay per PR
Best default for review workflows. Pair the live quote with a repo-level monthly cap and a clear pass / warn / block rule in the workflow.
AI-native indie builder
Pay per inspect
Best default for hosted CLI adoption. Keep the setup light, show the quote before spend, and let the user set a personal cap without reading infra docs.
Autonomous agent
Pay per action
Best default for automated systems. Pair per-action quotes with team or agent ceilings so unattended loops stay inside explicit spend policy.
Useful links
Stay inside persisted proof
These routes let maintainers move from runbook guidance back into live proof without hunting through internal docs.