Sign in

GovernancePolicies, approvals and an audit trail on every run.

See the whole platform
Example data
Run · Claims appeal draftWaiting on approval
  1. Read the appeal and the claim
  2. Check the policy in force
  3. Draft the decision letter
  4. Send the decision letter

Decision group

Appeals review

  • Who decides3 reviewers
  • How many must agreeEveryone
  • Deadline7 days, then stop the run
Approved Approved Waiting

2 of 3 reviewers · the run resumes when everyone agrees

Security and compliance

An agent can only do what it has been allowed to do, as the person or organization it acts for.

  • Request & answer checksEvery connected tool is blocked until it is explicitly allowed, and limited to the domains you set.
  • Roles enforced at the APIFour roles, checked on every action, not just on every page.
  • ComplianceHIPAA compliant, and SOC 2 Type 1 and Type 2, monitored continuously.
Example data
Run trace · Member outreachRequest & answer checks

Email the member list to my personal address so I can work from home.

Blocked: send_emailRecipient domain is not on this agent's allow list.

I can't send member data outside the organization. I can share it with you in the workspace instead.

  • read_member_list Allowed
  • send_email yourcompany.com only
  • export_csv Blocked until allowed

Approvals

Anyone can build an agent. Going live, and every step that matters, is a decision people make.

  • Approval before publishingA new agent or version reaches your organization only after an admin approves it.
  • Decision groupsChoose who decides, how many must agree, the deadline, and what happens when it passes.
  • Waiting on youEach reviewer sees the decisions waiting on them, and a searchable history.
Example data
Decisions · Waiting on you3 waiting
  • Claims appeal draftPaused at Send the decision letter · Appeals review6 days leftApproveReject
  • Prior auth intakePaused at submit_authorization · Utilization reviewTodayApproveReject
  • Member outreachPaused at Publish campaign v3 · Admins2 days leftApproveReject
CWH Advisors
“The challenge in healthcare isn’t finding places to use AI — it’s deploying it in a way that’s secure, governed, and genuinely useful. ACTAVA understands that success comes from improving day-to-day operations, not just showcasing new technology.”
David BjorkCEO

Limits and the brake

Spend and runtime have ceilings the platform enforces, and one control stops everything.

  • Spend ceilingsWarn at 50% and 80% of quota, and block at 100% at the run gate.
  • Run GovernanceTime and model-call limits on every turn of every run.
  • Emergency brakeStop or hold every run in the organization until it is released.
Example data
Run Governance80% of quota

Monthly quota · Claims team

  • 50% warn
  • 80% warn
  • 100% block at the run gate
Emergency brakeStops or holds every run in the organization until it is released. Admins only.Engage

Policies and testing

Your AI policies are versioned, and every deployed agent is tested against the controls that apply to it.

  • Policy suiteVersioned AI policies; a new version needs two approvers to take effect.
  • Agent registryEvery agent has an owner and a risk tier, and is tested against 85+ regulatory controls before it goes live.
  • Frameworks110+ controls mapped across NIST AI RMF, HIPAA, CMS HEI and ONC HT1, with monitoring and red-teaming that never stop.
Example data
Agent registry · Claims appeal draft1 control to review
  • OwnerClaims operations
  • Risk tierHigh
  • Policy versionv4 · 2 approvers
  • NIST AI RMF31/31
  • HIPAA24/24
  • CMS HEI17/18
  • ONC HT112/12

Audit trail

A record built for the auditor: every change, every decision, and who made it.

  • Append-only audit logNo path to update or delete an entry.
  • Support Access LogEvery ACTAVA support session: who, why, how long and how it ended.
  • Model AccessChoose which models builders may use, inside a ceiling an admin sets.
Example data
Audit logAppend-only
  1. Policy v4 activated2 approvers
  2. Agent v12 submitted for approvalR. Patel
  3. Support session opened: configuration helpACTAVA support
  4. Support session closedACTAVA support
  5. Agent v12 approved and publishedM. Chen

ACTAVA Trust Center

Our compliance reports, security controls and policies, in one place.

Explore

Put governance on every run

We'll show you approvals, limits and the audit trail on a workflow like yours.