Security

Security you can verify, not just take on trust.

Controls your administrators configure, cryptography you place in the workflow itself, and an execution record that shows what actually ran — not just a policy saying what should have.

Where we stand

Compliance, stated plainly

We would rather tell you exactly where each audit stands than show you a badge and let you assume the rest.

SOC 2

Approved for certification — certificate in the pipeline

The audit is closed and the certification is approved. The certificate itself is in the issuing pipeline and has not been issued yet.

ISO 27001

Approved for certification — certificate in the pipeline

Our information security management system has been audited against the standard and approved. The certificate is in the issuing pipeline.

Data protection

Controls built into the platform

Access control, obfuscation of sensitive fields, and a complete execution record are product features rather than add-ons. Bring your requirements and we will work through them with your team.

Public sector

GSA and Texas DIR

Public sector buyers can transact through the contract vehicles held by Ivoyant, our parent company.

Data protection

What happens to your data

Integration platforms hold credentials to everything you run. Here is how Koodisi treats them.

Secrets live in a key vault

Connection credentials and sensitive configuration are held in Azure Key Vault rather than in workflow definitions, so a secret is never a field someone can read off a canvas.

Encrypted in transit and at rest

Data moving between Koodisi and your systems, and data held by the platform, are both encrypted. Your workflows can add their own encryption on top.

Sensitive fields obfuscated

Fields you designate as sensitive are masked, so payloads stay readable enough to debug without exposing the values inside them.

Credentials managed centrally

Third-party account and source-control credentials are stored and rotated through the Credential Manager instead of being pasted into individual workflows.

Access control

Who can do what, and where

Access is configured by your administrators, at the scope the decision actually belongs to.

Single sign-on

Turn SSO on for the organization so users sign in with the company credentials they already have, and control it per user where you need an exception.

Permissions at two scopes

Permissions are granted at the organization level to apply across everything, or at the workspace level to stay contained. Each scope breaks down into its own subcategories.

Teams, not individuals

Group people into teams and grant access to the team. Members can be invited, updated, deactivated, and restored without unpicking individual grants.

Tenant isolation

Isolated runtimes, credentials, and audit trails per tenant, with access scoped to org, team, tenant, and environment.

Private workspaces

Mark a workspace private and it is visible only to the teams and users you add to it — useful when one group's integrations should not be discoverable by the rest of the org.

Named deployment approvers

Nominate who can approve a promotion. Nothing reaches production because someone had the button — it reaches production because a named approver signed off.

Encryption and certificates

Cryptography is a step, not a setting

Most platforms give you transport encryption and stop there. In Koodisi, encryption, signing, and signature verification are activities you place in a workflow — so a payload can be protected for the specific hop that needs it.

Certificates are configured with a public key path, private key path, and passphrase, then referenced by name from any workflow that needs them.

Crypto Utils

Encrypt, decrypt, hash, and sign data inside a workflow

Supports AES, RSA, and HMAC algorithms.

Encrypt

Secure a payload before it leaves your perimeter

Encrypts plain text or structured data using a configured certificate and algorithm — before sending to an external system or writing to a database.

Decrypt

Process sensitive inbound data safely

Decrypts Base64-encoded encrypted data using a configured certificate, so protected payloads from partners stay protected until the moment you need them.

Verify Signature

Prove an inbound message was not tampered with

Verifies a digital signature against a payload using a configured certificate, authenticating the sender and confirming message integrity.

Masking and selective logging

Debuggable without being exposed

The usual trade-off is between logs detailed enough to diagnose a failure and logs safe enough to keep. You decide field by field which side of that line each value sits on.

Log levelsDebugInfoWarnError

Mask the field, keep the record

The Data Formatter masks and restructures fields in flight, so an account number can be reduced to its last four digits before it is ever written anywhere.

Log at the level that suits the environment

Every Log step writes at a level you choose. Run Debug while you are building and drop to Warn in production without editing the workflow's logic.

Route messages to named loggers

Give a Log step a logger name and its output can be separated from everything else — so the noisy diagnostics from one integration stay out of the channel your on-call team watches.

Retention that matches the stakes

Execution metrics are retained longer in production than in non-production, and the window scales with your plan, so an investigation into a live incident still has the data it needs.

Governance and the audit record

The part that matters when someone asks what happened

Controls tell an auditor what should happen. The execution record tells them what did. Koodisi keeps both, because only one of them survives a real investigation.

Nothing reaches production unreviewed

Promotions run through governed deployments with named approvers, so the answer to 'who authorised this change' is recorded at the moment it happens rather than reconstructed afterwards.

Versioned, and reversible

Every promotion is versioned and can be rolled back. A bad change is a decision you can undo, not an incident you have to code your way out of.

Every execution is traceable

Executions, retries, and deployments are logged and traceable — the specific run that failed, the data it carried, and the step it failed on.

Metrics your team can actually read

Built on OpenTelemetry: average duration, error counts, request volume, and P50 and P99 latency, with execution flow diagrams showing where the time went.

For your security review

What your reviewers will ask for

Have a questionnaire to get through?

Send it over and we will fill it in with you, including anything on this page that your reviewers need in more detail.

Talk to us

FAQ

Questions we get in security reviews

If yours is not here, ask — we will answer it directly rather than pointing you at a portal.

Bring us your security review.

Send the questionnaire, the architecture questions, or the control matrix. We will work through it with you rather than pointing you at a portal.

Talk to us