canonical: https://jentic.com/apis/googleapis.com/workloadmanager

# Google Workload Manager API

The Google Workload Manager API automates the validation of enterprise workloads - SAP, Microsoft SQL Server, and other regulated deployments - against best-practice rules and recommendations on Google Cloud. Through evaluations and rules, teams discover deployed workload profiles, run on-demand or scheduled checks, and ingest insights for centralised reporting. The API exposes evaluations, executions, results, discovered profiles, and rules so engineering and SRE teams can shift workload compliance left into automation.

## For AI agents

Run best-practice validations against SAP, SQL Server, and other enterprise workloads on Google Cloud. List evaluations, trigger executions, and pull rule results for compliance reporting.

## Scope

Does not handle application code scanning, runtime workload execution, container image building, or non-Google-Cloud workloads - use for best-practice evaluation of enterprise workloads on Google Cloud only.

## Capabilities

- Create and list evaluations that target enterprise workloads such as SAP or SQL Server
- Trigger an evaluation execution on demand and track it via long-running operations
- Retrieve execution results enumerating which rules passed or failed for a workload
- List discovered workload profiles automatically detected in the project
- Write workload insights from external sources into Workload Manager via writeInsight

## Use cases

### Continuous SAP best-practice validation

SAP teams running on Google Cloud schedule evaluations to detect drift from Google's recommended SAP NetWeaver and HANA configurations. The API exposes both the evaluation definition and execution control surface so that runs can be triggered from CI on every infrastructure change. Rule results feed into the team's compliance dashboards and ticketing.

Example prompt: POST to /v1/{+name}/executions:run for an SAP evaluation and poll the returned operation until done, then list executions to retrieve the resulting rule pass/fail counts.

### SQL Server workload audits

Database administrators use Workload Manager to audit SQL Server deployments against Microsoft and Google best practices for HA, backup, and security configuration. Discovered profiles surface running instances automatically; evaluations encode the desired state. This produces an objective, repeatable audit trail in place of manual checklists.

Example prompt: List discoveredprofiles for type=SQL_SERVER, then for each profile create an evaluation execution and report any failed rules.

### Compliance ingestion from third-party scanners

Teams running additional scanners outside Google Cloud push workload-level insights into Workload Manager via /v1/{+location}/insights:writeInsight. This consolidates findings from Trivy, Wiz, or in-house tooling into the same evaluation-and-execution model the Cloud Console uses, giving security leadership a single view across native and third-party data.

Example prompt: POST to /v1/{+location}/insights:writeInsight with a sentenceFingerprint payload and verify the response contains a successful status.

### Agent-driven compliance reporting through Jentic

A compliance agent connected to Jentic can list evaluations, trigger executions, and assemble a report of failing rules without engineers writing client glue. Search Jentic for run a workload evaluation, load the executions:run operation, and execute it scoped to the right evaluation. Tokens stay isolated in your Jentic One instance.

Example prompt: Through Jentic, list evaluations, run each one via /v1/{+name}/executions:run, and produce a Markdown report grouping failed rules by workload type.

## Key endpoints

| Method | Path | Description |
| --- | --- | --- |
| GET | /v1/{+parent}/evaluations | List evaluations |
| POST | /v1/{+name}/executions:run | Trigger an evaluation execution |
| GET | /v1/{+parent}/discoveredprofiles | List discovered workload profiles |
| POST | /v1/{+location}/insights:writeInsight | Write a workload insight |
| POST | /v1/{+name}:cancel | Cancel a long-running operation |
| GET | /v1/{+name}/operations | List long-running operations |

## Key resources

- **evaluations** — Definitions that target a workload type with a set of rules
- **executions** — Individual runs of an evaluation
- **discoveredprofiles** — Automatically discovered workload profiles in a project
- **rules** — Best-practice rules executed during an evaluation
- **insights** — Third-party findings written into Workload Manager

## Why Jentic

- **Setup:** Wiring the Workload Manager API by hand means setting up OAuth 2.0 for your Google Cloud project, refreshing access tokens, and matching evaluation and insight calls against the workloadmanager.googleapis.com host. Through Jentic you install once, import the Workload Manager API from the API Directory, store the OAuth credential once, and your agent calls it.
- **Permission scoping:** The Workload Manager API puts the evaluation in the URL path (/v1/{name}/executions:run), so a rule can pin your agent to one evaluation: it can run that evaluation and read discovered profiles and nothing else. You choose the operations it may call, so cancelling executions is not included unless you add it.
- **Credential handling:** Your Google Cloud OAuth credential is stored once, encrypted, by your own Jentic One instance and injected at execution time. Its client secret and refresh token never enter the agent's prompt, logs, or context.
- **Discovery method:** Agents search Jentic by intent such as 'run a workload evaluation' or 'list discovered workload profiles', and Jentic returns the matching operation with its input schema so the agent calls the right endpoint without browsing the reference docs.

## Related APIs

- **Web Security Scanner API** — Application-layer vulnerability scanning for App Engine, Compute, and GKE
- **Cloud Asset Inventory API** — Inventories all GCP resources to scope evaluations
- **Cloud Monitoring API** — Surface evaluation failures as alerts in operational dashboards

## FAQ

### What authentication does the Workload Manager API use?

OAuth 2.0 (Oauth2 and Oauth2c schemes) scoped to https://www.googleapis.com/auth/cloud-platform. Through Jentic the OAuth credentials are stored encrypted and short-lived access tokens are exchanged at execution time.

### Which workload types does Workload Manager support?

The discoveredprofiles resource recognises SAP (NetWeaver and HANA), Microsoft SQL Server, and other enterprise workload types as Google adds rule packs. Evaluations target one workload type at a time. Check the discoveredprofiles list output for the supported types in your project.

### What are the rate limits for the Workload Manager API?

Per-project quotas for evaluation create and execution run calls are managed in the Cloud Console under APIs and Services. Discovery and read operations have generous defaults; the practical bound is how often you choose to run evaluations across the workload fleet.

### How do I trigger an evaluation through Jentic?

Search Jentic for run a workload evaluation, load the schema for POST /v1/{+name}/executions:run, and execute it with the evaluation name. The response is a long-running operation; poll /v1/{+name}/operations until done is true and then list executions to read rule results.

### Can I import findings from non-Google scanners?

Yes. POST /v1/{+location}/insights:writeInsight accepts a structured insight payload so you can centralise findings from Wiz, Trivy, or in-house scanners alongside Workload Manager's own evaluation results.

### Can I limit what my agent is allowed to do with the Workload Manager API?

Yes. Because you run Jentic One yourself, your own rules decide which Workload Manager operations and credentials the agent may use. Since the evaluation name sits in the URL path, such as /v1/{name}/executions:run, you can pin the agent to a single evaluation so it only runs that evaluation and lists discovered profiles. Operations you do not grant, like cancelling a long-running execution or writing insights, stay out of reach until you explicitly add them.