Which HIPAA technical safeguards Temporal helps with for ePHI workflows
Temporal's documented controls support several HIPAA technical safeguards in 45 CFR 164.312 for workflows that carry ePHI, but they do not complete compliance on their own. Covered entities and business associates must show that access, encryption, transmission security and audit activity are in place across the whole system, and Temporal covers only part of that system. The Security Rule reaches the workflow service, its persistence and visibility stores, workers, codec servers and the applications that start workflows, so the practical first step is to map each safeguard to a named component and a named owner before configuration begins.
For access control under 164.312(a)(1), the relevant settings are the Authorizer, which replaces the default noopAuthorizer that allows every API request, and the ClaimMapper, which maps JWT or certificate claims to namespace roles. Unique user identification under 164.312(a)(2)(i) is a required specification, so each human and service identity should trace to one account. For transmission security under 164.312(e)(1), mTLS with separate internode and frontend settings protects traffic between Temporal services and clients. The Web UI can require OIDC sign-in, but that controls access to the interface only and does not authorize calls to the Temporal Service, so the Authorizer must still be configured.
Encryption under 164.312(a)(2)(iv) is an addressable specification, so the team must document whether its custom codec and codec server meet the requirement and, if not, what equivalent measure applies. Payloads are encrypted by the codec before they reach the server, but workflow IDs, search attributes and other metadata remain visible to the server and visibility stores, so keep PHI out of them. Encryption at rest for the persistence database and visibility backend is a separate question to confirm with your storage and database configuration. For audit activity under 164.312(b), Temporal event history records workflow execution but does not by itself show who accessed ePHI through the Web UI or API, so confirm what your identity provider, API gateway and log pipeline capture and how long they retain it.
Several requirements sit outside the platform. Risk analysis under 164.308, written policies, workforce training and sanctions, physical safeguards under 164.310 and business associate contracts under 164.314(a)(2)(i) need organizational work. If you use Temporal Cloud or any hosted database, confirm the vendor's position on business associate terms before any ePHI enters the service. HHS states that no Security Rule standard requires certification, and no HIPAA certificate exists. DeployOpen plans and configures the self-hosted controls, prepares the system and evidence an assessor will request, and documents addressable decisions, but it does not issue audit reports, attestations or certifications. Independent assessors and the HHS Office for Civil Rights make the compliance judgment.
How each HIPAA technical requirement maps to Temporal controls
This table maps each HIPAA technical safeguard in scope to the Temporal control that supports it, the configuration you need, and whether the specification is required or addressable. Read it as a starting point for your risk analysis, since HIPAA defines no certification for compliance. The main takeaway is that Web UI sign-in and the Authorizer answer different questions, so both must be configured.
| Requirement | Temporal control and configuration | Addressable or required |
|---|---|---|
| 164.312(a)(1) Access control | Replace the default noopAuthorizer with an Authorizer, and map JWT or certificate claims to namespace roles with a ClaimMapper. | Standard. Its implementation specifications carry the required and addressable status. |
| 164.312(a)(2)(iv) Encryption and decryption | Encrypt workflow payloads with a custom codec, and serve decoding through a codec server only to authorized viewers. | Addressable. Document why it applies and any equivalent measure you choose. |
| 164.312(b) Audit controls | Use event history and visibility records, and confirm which access events your deployment logs for review. | Standard. It has no separate implementation specifications. |
| 164.312(e)(1) Transmission security | Enable mTLS with separate internode and frontend settings, and keep Temporal services off the public internet. | Standard. Its encryption specification (e)(2)(ii) is addressable. |
| Web UI access point | Set TEMPORAL_AUTH_ENABLED with OIDC sign-in. This controls interface access only, not Service authorization. | Falls under access control. The Authorizer must still enforce namespace permissions. |
| Documenting addressable items | Record the risk assessment, the decision for each addressable specification, and any alternative measure in your Security Rule documentation. | Entity must document its reasoning for each addressable item. |
How a codec server encrypts ePHI in workflow payloads
The codec is where ePHI in a Temporal workflow payload becomes ciphertext, and it is also where that ciphertext can become readable again. Payloads pass through a custom codec before they leave the client, and the Web UI can decode them through a codec server endpoint. That makes encryption keys and decryption rights a design decision, not a configuration detail. Under 45 CFR 164.312(a)(2)(iv), encryption and decryption is an addressable implementation specification. You must assess whether it is reasonable and appropriate for your workflows, and if you do not implement it as written, document why and adopt an equivalent measure. Record that assessment before the codec is built.
Key management comes first. Decide where keys are generated, stored and rotated, and whether they sit in a managed key service in your cloud account, a key manager in your data center, or a secrets platform you operate. Use separate keys for each environment and each data class so a development codec cannot decrypt production payloads. Confirm how rotation affects long-running workflows that still hold payloads encrypted under an older key, and how keys are backed up and recovered. Ask your team or your implementation partner to document which key systems the codec implementation supports and who holds the key administration role.
Next, map which identities may decrypt. Workers run workflow and activity code, so they need the codec to read their inputs. Operators, support staff and auditors need decoded data only for specific tasks, and the Web UI decodes on behalf of whoever is signed in. List each identity type: worker service accounts, operators, support, auditors and any automation. Then use a ClaimMapper to map JWT or certificate claims to namespace roles, and replace the default noopAuthorizer, which allows every API request, with an Authorizer plugin. The TEMPORAL_AUTH_ENABLED setting controls sign-in to the interface but not authorization against the Temporal Service, so a signed-in user still needs Authorizer rules that match the roles you defined.
The codec server is a high-value target because it decrypts on request. Keep it on hosts that are not reachable from the public internet, restrict network access to trusted internal networks, and require authentication before any decode request is served. Use mTLS for internode and frontend traffic and confirm that the codec endpoint is covered by the same trust model. Log decode requests so access can be reviewed against 45 CFR 164.312(b). Then confirm what the codec does not cover. The supplied facts do not describe encryption at rest for the persistence database or the visibility store. Confirm how your chosen Cassandra, PostgreSQL or MySQL persistence and Elasticsearch, OpenSearch or SQL visibility store encrypt data at rest, and whether search attributes or memos in clear text could carry ePHI.
Configuring mTLS, the Authorizer and Web UI sign-in in the right order
This sequence applies the controls in the order Temporal documents them, so each layer is in place before the next one depends on it. The result is a self-hosted cluster where network placement, transport encryption, authorization, sign-in and payload encryption are each verified before any ePHI enters a workflow.
Place Temporal services on private networks
Your platform or infrastructure team places Temporal Server services on hosts that are unreachable from the public internet, with network access limited to trusted internal networks, as the Temporal deployment documentation describes. Before installing, confirm the Helm chart version, because chart versions below 0.73.1 do not work with server images 1.30 and later. Plan server upgrades one minor version at a time, moving to the latest patch of the current minor first, and run schema upgrades before server upgrades when the release notes call for them. This step is done when a scan from outside your trusted networks cannot reach the frontend, internode ports or persistence database, and the cluster starts on the version you chose.
Enable mTLS for internode and frontend traffic
Configure mutual TLS separately for internode traffic between Temporal services and for frontend traffic from clients and workers, so both sides present certificates. This supports the transmission security standard in 45 CFR 164.312(e)(1), which asks for protection against unauthorized access to ePHI sent over electronic networks. Issue certificates from a certificate authority your security team controls, and record who owns renewal and rotation. This step is done when a client without a trusted certificate is refused at the frontend, and a worker with a valid certificate connects and completes a test workflow that contains no real ePHI.
Replace the noopAuthorizer with ClaimMapper and Authorizer
Without an Authorizer, self-hosted Temporal uses the default noopAuthorizer, which allows every API request with no authentication or access control. Configure a ClaimMapper that maps JWT or certificate claims to namespace roles, then enable an Authorizer plugin that enforces those roles on each request. Your engineers or implementation partner write and test the Authorizer logic against the claims your identity provider actually issues. This step is done when a principal with no role is denied namespace operations, and a read-only principal cannot start or terminate workflows. Keep the test results as evidence for the access control standard in 45 CFR 164.312(a)(1).
Design namespace roles around care team access
Map each namespace role to a documented need to know, starting from the workflows that will carry ePHI and the teams that operate them. Separate namespaces where data sensitivity or ownership differs, and avoid broad grants that give one principal access to every namespace. Ask the data owner to approve the role matrix before it reaches the ClaimMapper configuration, so the policy is set by the people accountable for the data. This step is done when the approved matrix matches what the ClaimMapper produces in a test environment, and each role has a named owner who reviews membership on the schedule your policy sets.
Configure OIDC sign-in for the Web UI
Enable OIDC sign-in for the Temporal Web UI with the TEMPORAL_AUTH_ENABLED setting and the identity provider settings for your version, using the Web UI configuration reference as the source for exact variable names. Sign-in controls who can reach the interface, while authorization against the Temporal Service still comes from the Authorizer you configured in the previous step. Confirm with your identity provider team that multi-factor authentication is enforced for users who can open ePHI workflows. This step is done when an unauthenticated browser is redirected to the identity provider, and a signed-in user without a namespace role sees no workflow data.
Deploy the codec server and connect the Web UI
Implement a custom payload codec that encrypts workflow payloads before they are stored, then deploy the codec server that decrypts them for authorized viewers in the Web UI. Keep encryption keys in a key management service you control, and restrict the codec endpoint to the same authorized users as the Web UI. Payload encryption is an addressable implementation specification under 45 CFR 164.312(a)(2)(iv), so document why it is reasonable and appropriate for your design. This step is done when the raw payload in workflow history is unreadable, an authorized user sees the decoded value in the Web UI, and an unauthorized user does not.
Verify each control and retain the evidence
Test each control in sequence against a non-production copy of the configuration: network placement, mTLS on both traffic types, Authorizer decisions, Web UI sign-in, and codec output. Confirm that audit logs of authentication and administrative activity are kept and reviewed, which supports the activity recording standard in 45 CFR 164.312(b). Store test results, configuration files and role approvals together so an assessor can trace each safeguard to its evidence. HHS does not certify HIPAA compliance, so an independent assessor or your own compliance function makes the determination. This step is done when every control has a dated test record and an owner who can explain it.
What audit evidence a Temporal ePHI configuration produces
Each card describes a document or record an assessor may request when reviewing a self-hosted Temporal deployment that carries ePHI in workflow payloads. Together they show how access, encryption, logging and change control were configured and how they stay accurate over time. Keep each one current through your normal change management process.
mTLS and Authorizer configuration
The deployment manifests and configuration files that set mTLS for internode and frontend traffic and load the Authorizer plugin. An assessor will look for proof that the default noopAuthorizer, which allows every API request, was replaced in each environment that carries ePHI. Keep these files in version control, tie each release to a reviewed commit, and record which certificate authority issues service and client certificates. Your risk analysis determines which transmission and access settings must be documented.
ClaimMapper and namespace role mappings
The rules that map JWT or certificate claims to namespace roles, along with the approval record for each role. For ePHI namespaces, the assessor will ask who can read workflow histories, start executions and administer the cluster, and how those permissions match the access rights your access control procedures grant. Export the mapping on every change, compare it against identity provider groups on a fixed schedule, and remove roles when duties change. Your risk analysis decides how granular the roles must be.
Codec server design and key custody records
The design of the codec server that encrypts and decrypts payloads, including which namespaces it serves, which services call it, and how it authenticates those callers. Key custody records show where encryption keys are generated, who can access them, how they rotate, and how payloads written under a retired key remain readable. Update the design document and key inventory with every codec change. Confirm with your security team how the encryption specification in 164.312(a)(2)(iv) is documented, because the entity decides whether its approach is reasonable and appropriate.
Web UI OIDC configuration
The TEMPORAL_AUTH_ENABLED settings, the OIDC client registration and redirect configuration for the Web UI, and the codec endpoint the UI calls to decode payloads for operators. Assessors will check that interface sign-in is enforced and that Web UI access is documented as separate from API authorization, which the Web UI setting does not provide. Record the identity provider, the client scopes and who approved them. Review the configuration after each Web UI upgrade and after identity provider changes.
Access and audit activity records
Logs from the Temporal frontend, Authorizer decisions, the codec server and the Web UI, retained and reviewed in a way that supports 164.312(b), which requires mechanisms that record and examine activity in systems containing ePHI. Check that logs show who accessed which namespace and when a decoded payload was requested. Confirm which events your Temporal version emits, where logs are stored, who can alter them, and who reviews them. Retention periods come from your policies and risk analysis.
Upgrade and test records
Records of each Temporal Server upgrade, made one minor version at a time with schema upgrades completed first where release notes require them, plus test results for mTLS, Authorizer denials, codec round trips and Web UI sign-in. Keep the upgrade plan, the server version tested, and the rollback decision together with each change. Rerun the test set after every upgrade. Your risk analysis determines how often testing repeats and which results your program accepts as evidence.
What the Security Rule still asks for beyond Temporal configuration
The HIPAA Security Rule applies to covered entities (health plans, health care clearinghouses and covered health care providers) and to their business associates, for all electronic protected health information they create, receive, maintain or transmit. Temporal configuration covers only part of that obligation. The technical safeguards in 45 CFR 164.312 address access, audit controls, integrity, authentication and transmission security for the Temporal services, but the administrative and physical safeguards in 164.308 and 164.310 still need owners, documented procedures and evidence. Before you design the deployment, confirm with your compliance lead or counsel which entity type you are and whether your workflows make you a business associate of someone else.
Most of the Temporal-related specifications are addressable, not required. For an addressable specification, the entity must assess whether the measure is reasonable and appropriate in its environment. If it is not, the entity must document why and implement an equivalent alternative measure where that is reasonable and appropriate (45 CFR 164.306(d)). Payload encryption under 164.312(a)(2)(iv) is one example. Your written assessment should state how the codec and key arrangement is chosen, who holds decryption rights, and what happens to data that reaches the visibility store, logs or the Web UI in plain form. Record the reasoning even when the outcome is the control you would have chosen anyway.
If a vendor, hosting partner, managed database provider or implementation partner creates, receives, maintains or transmits ePHI for you, 45 CFR 164.314(a)(2)(i) requires a business associate contract. That contract must require the business associate to comply with the Security Rule, flow the same terms to subcontractors that handle ePHI, and report security incidents, including breaches of unsecured PHI. Ask each party whether anyone on their side can see decrypted workflow payloads, who can reach the Temporal namespaces, and where backups and support artifacts are stored. Have counsel review whether the contract language matches those answers.
HHS states that no Security Rule standard requires an entity to certify compliance and that it does not endorse or recognize private certifications, which also do not prevent HHS from later finding a violation (HHS FAQ 2003). Treat any vendor claim of HIPAA certification as a statement to verify, not as evidence. The HHS Office for Civil Rights enforces the rule through complaint investigations, compliance reviews and audits. Its 2024 to 2025 HIPAA audits review 50 covered entities' and business associates' compliance with provisions most relevant to hacking and ransomware. The inflation-adjusted civil money penalty for penalties assessed on or after January 28, 2026 ranges from $145 to $2,190,294 per violation, with a calendar-year cap of $2,190,294 (45 CFR 102.3, 91 FR 3666).
NIST SP 800-66 Rev. 2 (February 2024) is a NIST resource guide for implementing the Security Rule. It gives key activities and sample questions for each standard, and HHS lists it as informational rather than legally binding. It is useful for building the risk analysis and test plan for a Temporal deployment. HHS proposed a major Security Rule revision on January 6, 2025 (90 FR 898, RIN 0945-AA22). As of October 2026 no final rule has been published, and the latest Unified Agenda projects final action for July 2027. Build the current rule into your assessment now, and ask counsel how a final rule would change the documentation you need to keep.
What Temporal does not cover for HIPAA
Self-hosted Temporal can support the technical safeguards for ePHI in workflow payloads, but it does not make a deployment HIPAA compliant. The cards below list the obligations that remain with your organization, its vendors and its operations team, along with what to confirm for each one before production use.
No HIPAA certification or attestation
HHS states that no Security Rule standard requires certification, and that private certifications do not stop HHS from later finding a violation. Temporal does not issue a HIPAA certification or attestation for its server, so no vendor document can stand in for your own compliance evidence. Ask Temporal for its security documentation and confirm exactly which statements it makes about HIPAA, then record how your own controls satisfy each applicable Security Rule standard.
Risk analysis, policies and workforce training
The administrative safeguards in 45 CFR 164.308 cover the risk analysis, risk management, information system activity review, written policies, workforce security and sanctions. These are entity obligations that no Temporal configuration satisfies. Confirm who performs the risk analysis for the workflow platform, who approves access to the Web UI and namespaces, how training is delivered, and which sanctions policy applies to staff who misuse access to payload data.
Physical and facility safeguards
The physical safeguards in 45 CFR 164.310 cover facility access controls, workstation use and security, and device and media controls. Temporal services, databases and worker hosts run on machines you or your hosting provider control, so these safeguards apply to every data center, rack, cloud region and laptop that holds ePHI or its keys. Confirm the physical controls of each hosting location, how workstations that open the Web UI are secured, and how disposal and reuse of storage media are handled.
Business associate contracts and Temporal Cloud
A business associate contract under 45 CFR 164.314(a)(2)(i) must require the vendor to follow the Security Rule, flow the same terms to subcontractors that handle ePHI, and report security incidents. Any hosting provider, database provider or Temporal Cloud that can access ePHI needs a contract. The supplied facts do not establish whether Temporal Cloud is suitable for ePHI or will sign a business associate agreement, so ask Temporal in writing before any ePHI enters a managed namespace.
Encryption at rest for persistence and visibility
Payload encryption with a codec server protects workflow payloads, but the persistence database and the visibility store can hold ePHI in other forms, such as search attributes and execution history. Encryption at rest for these stores depends on the database and search engine you select, along with their storage configuration and key management. Confirm the encryption at rest options for your chosen PostgreSQL, MySQL, Elasticsearch, OpenSearch or other supported store, and document where the keys are held.
Backup and disaster recovery testing
The contingency plan standards in 45 CFR 164.308(a)(7) and the data backup and disaster recovery specifications require tested recovery, not only configured backups. Temporal persistence and visibility stores must be backed up together, and encrypted backups need the same access controls as live data. Confirm backup frequency, restore procedures, retention periods and the recovery tests you will run, and record who approves restoring ePHI into a different environment.
Questions teams ask before running ePHI on Temporal
How much does a self-hosted Temporal implementation for ePHI cost?
The cost depends on the written scope, and DeployOpen does not publish a price list, so a quote follows from the agreed deliverables. Expect several cost components: DeployOpen's fixed-scope fee for design, build, testing and evidence; your own cloud or data center capacity for the Temporal services, database and visibility store; your staff time for identity, network and approvals; and an optional retainer for upgrades and support after go-live.
Cost drivers: the number of namespaces, the codec and key design, Authorizer and ClaimMapper integration with your identity provider, the database and visibility choices, testing depth, and how much evidence collection your assessors expect.
For comparison only, Temporal Cloud pay-as-you-go pricing starts at 50 US dollars per million actions, plus storage charges, according to Temporal's pricing page. Ask any partner to itemize one-time and recurring costs in the quote.
How long does an ePHI Temporal deployment take to deliver?
Duration is set by the written scope, and no fixed timeline applies to every deployment. The main variables are the number of namespaces and workflow types that carry ePHI, how quickly your identity provider can issue the claims that the ClaimMapper needs, whether a codec server must be built and reviewed, and how long your security and compliance reviewers take to accept the evidence.
Infrastructure lead time also matters. Your team must provision hosts that are not reachable from the public internet, plus a supported database and visibility store. Confirm each milestone, its owner and its acceptance criteria in the statement of work before work starts, and ask for the dependencies on your side to be listed explicitly.
What does our team need to provide for an outside implementation?
You provide access, decisions and people. Plan for a named technical owner who can approve architecture choices, access to your cloud accounts or data center network, and a route to your identity provider administrators for OIDC sign-in, mTLS certificate issuance and role mapping.
You also provide the data classification for each workflow, the list of systems that send or receive ePHI, and the policies that your compliance lead already uses for access reviews, logging and key management. Your security team should confirm where encryption keys will be held and who may administer them. Ask the partner to list each input as a dependency in the written scope so that delays are visible early.
Can our in-house platform team run this without an outside partner?
Yes, a capable in-house team can build this, but the work requires skills across distributed workflow design, Kubernetes or host operations, PKI, identity federation and security evidence. Temporal's documentation covers deployment, persistence, visibility, mTLS, Authorizer, ClaimMapper and codec options, so the knowledge is available. The harder part is making consistent decisions across all of them and documenting why each one meets the relevant safeguard.
Consider three questions before deciding: whether your team has run a stateful distributed service in production, whether someone can own the upgrade sequence (one minor version at a time, with schema upgrades first when the release notes call for them), and whether your assessors will accept internal testing as evidence. If the answer to any of these is uncertain, a fixed-scope engagement can cover the gap.
What happens if we leave the default Temporal authorization in place?
If no Authorizer is configured, self-hosted Temporal uses the default noopAuthorizer, which allows every API request with no authentication or access control. For ePHI, this default fails the access requirement in 45 CFR 164.312(a)(1), which calls for technical controls so that only persons or software programs with granted access rights can reach ePHI.
An Authorizer plugin, paired with a ClaimMapper that maps JWT or certificate claims to namespace roles, is the documented way to enforce namespace-level access. Web UI sign-in through OIDC controls who can open the interface, but it does not replace authorization against the Temporal Service. Treat these as separate controls, test each one with denied and allowed identities, and keep the test results as evidence.
Does a payload codec server meet the HIPAA encryption requirement?
A codec that encrypts workflow payloads supports the mechanism in 45 CFR 164.312(a)(2)(iv), but it does not settle the requirement alone. The encryption and decryption specification is addressable, so your organization must assess whether it is reasonable and appropriate, and document the decision and any equivalent measure.
Check these points in your design: who can call the codec server endpoint and decode payloads in the Web UI, how encryption keys are generated, stored, rotated and revoked, whether transport between components uses mTLS with separate internode and frontend settings, and how the database and visibility store protect data at rest. Payloads can also leak into logs or search attributes, so review what the visibility store indexes before you rely on payload encryption alone.
Does following this page make our Temporal deployment HIPAA compliant?
No. HIPAA has no certification, and HHS states that no Security Rule standard requires an entity to certify compliance and that HHS does not endorse or recognize private certifications. Compliance is an obligation of each covered entity or business associate, and the HHS Office for Civil Rights enforces the rule through complaints, compliance reviews and audits.
This page describes how specific Temporal controls can support technical safeguards in 45 CFR 164.312, including access control, audit controls, encryption and transmission security. Your organization still decides how each addressable specification is handled, runs the risk analysis and documents its policies. Independent auditors and certification bodies issue any attestations. DeployOpen prepares systems, controls and evidence for those assessments.
How DeployOpen works on this
DeployOpen plans, implements and documents self-hosted Temporal inside your own infrastructure, so ePHI-carrying workflows run under controls your team can inspect and maintain. Each engagement starts from a written, fixed scope that names the namespaces, workflows, codec, identity and evidence that will be delivered.
How we scope
- Inventory which workflows carry ePHI in inputs, results, signals and failure messages, and which namespaces they run in.
- Agree in writing which controls are in scope: payload encryption with a codec server, mTLS, Authorizer, Web UI sign-in and logging.
- Confirm the hosting, persistence and visibility stores and whether they are also in scope for encryption and access controls.
How we build
- Design the codec server and key handling so workers and the Web UI decode payloads only for authorized principals.
- Configure mTLS separately for internode and frontend traffic, and map certificate or JWT claims to namespace roles with a ClaimMapper.
- Replace the default noopAuthorizer with a tested Authorizer and verify denied calls, plus OIDC sign-in for the Web UI.
How we hand over
- Deliver configuration as code, with a control-to-requirement map that links each setting to the HIPAA technical safeguard it supports.
- Document the evidence your assessors will ask for, such as access logs, authorization test results and codec configuration records.
- Walk your security and platform teams through key rotation, role assignments and the steps to change the codec or Authorizer.
How we support
- Offer ongoing support, upgrades and continuous testing as a retainer after the fixed-scope delivery ends.
- Plan Temporal Server upgrades one minor version at a time and re-run authorization and encryption checks after each change.
- Confirm with your compliance and legal advisers which HIPAA obligations apply to your organization, since DeployOpen prepares controls and evidence but does not issue certifications.

