Skip to main content

Conscience · Policy & governance

Every AI action gets an audit row. Every denial cites a rule.

Conscience defines what your agents are allowed to do, records what they did, and emits tamper-evident receipts every paid op can be audited against. Built for regulated industries that need the AI to be auditable, not just productive.

Available from Starter — $19 CAD/moRead the docs

What it does

Declarative policy packs

Tenant-scoped rule sets in plain language. SYSTEM_RULES always load first; tenant rules layer on top. Update without redeploying.

Fail-closed by default

If no policy explicitly allows an action, it's denied. Every denial emits an audit row with the rule id that fired. No silent skips.

Hash-chained receipts

Every cloud op produces a receipt linked to the prior one. Offline-verifiable. Tampering breaks the chain in a detectable way.

Jurisdictional templates

Pre-built packs for GDPR, HIPAA, PIPEDA, SOC 2, ISO 27001, and Canadian + US tax handling. Adopt or fork.

When to reach for it

  • You operate in a regulated industry (healthcare, finance, public sector) and need an auditable trail.
  • You want to constrain agent behaviour declaratively rather than with prompt tricks.
  • You're preparing for a SOC 2 / ISO / HIPAA audit and need receipts your auditor can verify.
  • You're wiring multiple teams into shared AI agents and need per-team rule packs.

Deep dives

Pricing

This product is included starting from the Starter tier ($19 CAD/mo). Every higher tier includes it automatically. See the tier & product mapping for the full picture.

Read more