We know our customers get through deployment and then get stuck. Governance workflows are running, but the data behind them has only moved one way: evidence could be reviewed, not written back; use cases could be read, but not extended with fields specific to your organization; audit activity existed, but pulling a clean trail out meant asking someone for an export.
Today that changes. This release makes governance evidence a true two-way system: the same foundation that powers manual review inside Credo AI is now available to the integrations, automations, and assistants your teams already build with.
What is Credo AI's SDK 1.2?
Four capabilities, one API surface:
1. Evidence, read and write
Submit and retrieve answers against questionnaires (use case and model) and control evidence requirements across every supported answer type: text, number, file, date, single/multi-select, metric, table, and more.
2. Custom use case fields
Extend the use case model with fields specific to your organization business unit, risk tier, deployment region, or anything else your governance program tracks, and use them to filter and drive workflows.
3. Audit log export
Pull governance activity out of your tenant on demand, with column selection and filtering, in JSON or CSV.
4. Programmatic report generation
List and generate report templates through the API: review packages and executive summaries without leaving your existing tooling.
Why it matters?
A one-way API isn't a governance system: Reading governance data is useful. Writing it back is what makes an API a governance system instead of a reporting layer. A team migrating their AI inventory out of Snowflake or a CMDB needs to map their own fields onto the use case model, then patch in evidence without re-entering it by hand. A compliance team pushing evidence from a GRC tool needs it to land as real control evidence, not a manual copy-paste. An audit team needs governance activity in the SIEM they already watch, not a one-off export request. This release is built for exactly those flows.
This is one piece of Credo AI's move toward a programmable governance platform: the same Integration Service behind this release is what our SDKs, and eventually our MCP server, are built on.
Who is it for?
Built for the teams closing the loop:
1. AI Governance Teams
Query evidence completion directly instead of checking each use case by hand, which is the groundwork for automated pre-review checks.
2. Business & Product Teams
Push evidence from the systems you already use straight into Credo AI, instead of re-keying it into a questionnaire.
3. Compliance & Audit Teams
Bring governance activity into your existing SIEM/GRC tooling on your own schedule, in the format it expects.
4. Platform & Integration Teams
Map your own fields from Snowflake, a CMDB, or wherever your AI inventory already lives — onto the use case model, without waiting on a schema change from Credo AI.
Security and Permissions
Nothing new to configure
Evidence, field, and audit operations run through Credo AI's existing Roles and Permissions model, including any custom roles your organization has configured — consistent with how every other part of the platform already works. If you'd like this validated against a specific custom role before you build against it, your account team can confirm it for your tenant.
WHAT'S NEXT
Still rolling out. More updates to come including:
1. Simpler SDK patterns for working with evidence and questionnaires, so you're not assembling raw payloads by hand.
2. File attachment support extending to questionnaire answers, matching what's already available for control evidence.
3. Questionnaire authoring via API — create and version sections and questions, not just answer existing ones.
Get Started
Put this to work in your tenant: If you're already integrating with Credo AI, this is available through the same Integration Service and SDK you're using today. If you're not yet, your account team can walk through what a first evidence-sync integration looks like for your stack.
Talk to your Credo AI account team for more.
DISCLAIMER. The information we provide here is for informational purposes only and is not intended in any way to represent legal advice or a legal opinion that you can rely on. It is your sole responsibility to consult an attorney to resolve any legal issues related to this information.
