Core capabilityApplication connectorsUse supported connector operations to connect business applications.
4 points maximum | Supported4 / 4 pts Plans include application integrations. Official documentation · Oct 15, 2026Source ↗ |
|---|
Core capabilityCustom APIs or webhooksConnect a custom HTTP API, webhook or custom connector.
4 points maximum | Not verified0–4 / 4 pts The reviewed documentation does not establish custom apis or webhooks at this feature level. This is an evidence gap, not a finding that the capability is absent. Not verified · Oct 15, 2026Source ↗ |
|---|
Depth checkConnection identity and access scopeConfigure or inspect the identity/credentials and scope used by a connected application.
5.6 points maximumAsk for a demonstrationShow which account a workflow uses and which data/actions its connection can access. A scenario to run with the supplier; not a completed product test. | Not verified0–5.6 / 5.6 pts The reviewed evidence does not establish this advanced behavior: connection identity and access scope. Request the demonstration above; absence of public evidence does not establish absence of the feature. Not verified · Oct 15, 2026Source ↗ |
|---|
Depth checkAuthenticated webhook or API triggerProtect an inbound webhook/API trigger with authentication or documented verification controls.
6.4 points maximumAsk for a demonstrationCall a test endpoint without valid credentials and show it being rejected. A scenario to run with the supplier; not a completed product test. | Not verified0–6.4 / 6.4 pts The reviewed evidence does not establish this advanced behavior: authenticated webhook or api trigger. Request the demonstration above; absence of public evidence does not establish absence of the feature. Not verified · Oct 15, 2026Source ↗ |
|---|
Core capabilityConditional workflow pathsBranch or filter workflow execution based on conditions.
4 points maximum | Supported4 / 4 pts Multi-step workflow logic is documented. Official documentation · Oct 15, 2026Source ↗ |
|---|
Core capabilityData transformation stepsMap, format or transform data between workflow steps.
4 points maximum | Not verified0–4 / 4 pts The reviewed documentation does not establish data transformation steps at this feature level. This is an evidence gap, not a finding that the capability is absent. Not verified · Oct 15, 2026Source ↗ |
|---|
Depth checkIteration with conditional pathsProcess a list with loops/iteration and branch according to each item’s result.
5.6 points maximumAsk for a demonstrationProcess three records, route one exception separately and show the per-record results. A scenario to run with the supplier; not a completed product test. | Not verified0–5.6 / 5.6 pts The reviewed evidence does not establish this advanced behavior: iteration with conditional paths. Request the demonstration above; absence of public evidence does not establish absence of the feature. Not verified · Oct 15, 2026Source ↗ |
|---|
Depth checkExplicit error branch or fallbackDefine what should happen when an intermediate action fails.
6.4 points maximumAsk for a demonstrationFail the second step and show the configured fallback rather than a silent incomplete run. A scenario to run with the supplier; not a completed product test. | Not verified0–6.4 / 6.4 pts The reviewed evidence does not establish this advanced behavior: explicit error branch or fallback. Request the demonstration above; absence of public evidence does not establish absence of the feature. Not verified · Oct 15, 2026Source ↗ |
|---|
Core capabilityExecution history and monitoringInspect workflow runs, job history or execution status.
3.2 points maximum | Supported3.2 / 3.2 pts Execution history and insights are plan-dependent. Official documentation · Oct 15, 2026Source ↗ |
|---|
Core capabilityFailure handling or replayHandle errors or retry/replay a failed execution.
3.2 points maximum | Not verified0–3.2 / 3.2 pts The reviewed documentation does not establish failure handling or replay at this feature level. This is an evidence gap, not a finding that the capability is absent. Not verified · Oct 15, 2026Source ↗ |
|---|
Depth checkPer-step input/output diagnosticsInspect individual step results and errors in a workflow execution.
4.8 points maximumAsk for a demonstrationOpen a failed run and show which step failed and the input/output relevant to repair. A scenario to run with the supplier; not a completed product test. | Not verified0–4.8 / 4.8 pts The reviewed evidence does not establish this advanced behavior: per-step input/output diagnostics. Request the demonstration above; absence of public evidence does not establish absence of the feature. Not verified · Oct 15, 2026Source ↗ |
|---|
Depth checkControlled retry or failed-run replayConfigure or initiate retries/replay with visible execution boundaries.
4.8 points maximumAsk for a demonstrationRetry the failed work and explain which completed actions will or will not run again. A scenario to run with the supplier; not a completed product test. | Supported4.8 / 4.8 pts Failed runs can retry with the saved workflow and previous execution data. Official documentation · Oct 15, 2026Source ↗ |
|---|
Core capabilityWorkflow or team permissionsLimit access to workflows, teams or shared connections.
2.4 points maximum | Supported2.4 / 2.4 pts SSO and administrative controls depend on edition. Official documentation · Oct 15, 2026Source ↗ |
|---|
Core capabilityDeployment or connection policy controlsUse environments, managed secrets, connector policies or comparable deployment controls.
2.4 points maximum | Supported2.4 / 2.4 pts Environments and external secrets depend on edition. Official documentation · Oct 15, 2026Source ↗ |
|---|
Depth checkWorkflow version or change historyInspect or restore earlier workflow definitions or a change history.
3.2 points maximumAsk for a demonstrationChange a workflow, then show the previous version and restoration/audit options. A scenario to run with the supplier; not a completed product test. | Supported3.2 / 3.2 pts Source-control environments preserve workflow changes in a private repository. Official documentation · Oct 15, 2026Source ↗ |
|---|
Depth checkRestricted shared credentials or secretsControl which collaborators can use or administer shared credentials/secrets.
4 points maximumAsk for a demonstrationGive a colleague workflow access while showing the boundary around connection credentials. A scenario to run with the supplier; not a completed product test. | Not verified0–4 / 4 pts The reviewed evidence does not establish this advanced behavior: restricted shared credentials or secrets. Request the demonstration above; absence of public evidence does not establish absence of the feature. Not verified · Oct 15, 2026Source ↗ |
|---|
Core capabilityAI workflow-building assistanceUse natural language or AI assistance to construct or change a workflow.
2.4 points maximum | Supported2.4 / 2.4 pts Workflow Assistant use has hosted-plan allowances. Official documentation · Oct 15, 2026Source ↗ |
|---|
Core capabilityAI or agent steps at runtimeRun a model, agent or tool-enabled AI step inside an operational workflow.
2.4 points maximum | Supported2.4 / 2.4 pts AI nodes and agents run with provider-specific model charges and limits. Official documentation · Oct 15, 2026Source ↗ |
|---|
Depth checkAI step uses configured business contextProvide an AI step with explicit workflow data, knowledge or connected business records.
3.2 points maximumAsk for a demonstrationShow the exact source records passed into an AI step for a sample execution. A scenario to run with the supplier; not a completed product test. | Not verified0–3.2 / 3.2 pts The reviewed evidence does not establish this advanced behavior: ai step uses configured business context. Request the demonstration above; absence of public evidence does not establish absence of the feature. Not verified · Oct 15, 2026Source ↗ |
|---|
Depth checkHuman approval before AI actionsInsert an approval or review boundary before AI-generated output triggers a consequential action.
4 points maximumAsk for a demonstrationLet AI propose a record change and show the run waiting for human approval. A scenario to run with the supplier; not a completed product test. | Supported4 / 4 pts The Gmail approval operation pauses execution while waiting for a human response. Official documentation · Oct 15, 2026Source ↗ |
|---|
Depth checkDetect prompt-injection attempts before agent tool useDemonstrate: Detect prompt-injection attempts before agent tool use. State the plan, limits and behaviour on failure. Use a sample event, a repeated event and a failing destination.
2.5 points maximumAsk for a demonstrationDemonstrate: Detect prompt-injection attempts before agent tool use. State the plan, limits and behaviour on failure. Use a sample event, a repeated event and a failing destination. A scenario to run with the supplier; not a completed product test. | Not verified0–2.5 / 2.5 pts This specific check has not been verified. Missing evidence does not mean the capability is absent. Ask for a demonstration in the proposed plan. Not verified · Oct 21, 2026Source ↗ |
|---|
Depth checkOmit sensitive payloads from stored execution logsDemonstrate: Omit sensitive payloads from stored execution logs. State the plan, limits and behaviour on failure. Use a sample event, a repeated event and a failing destination.
2.5 points maximumAsk for a demonstrationDemonstrate: Omit sensitive payloads from stored execution logs. State the plan, limits and behaviour on failure. Use a sample event, a repeated event and a failing destination. A scenario to run with the supplier; not a completed product test. | Not verified0–2.5 / 2.5 pts This specific check has not been verified. Missing evidence does not mean the capability is absent. Ask for a demonstration in the proposed plan. Not verified · Oct 21, 2026Source ↗ |
|---|
Depth checkPreserve event order across successive executionsDemonstrate: Preserve event order across successive executions. State the plan, limits and behaviour on failure. Use a sample event, a repeated event and a failing destination.
2.5 points maximumAsk for a demonstrationDemonstrate: Preserve event order across successive executions. State the plan, limits and behaviour on failure. Use a sample event, a repeated event and a failing destination. A scenario to run with the supplier; not a completed product test. | Not verified0–2.5 / 2.5 pts This specific check has not been verified. Missing evidence does not mean the capability is absent. Ask for a demonstration in the proposed plan. Not verified · Oct 21, 2026Source ↗ |
|---|
Depth checkChoose pause or discard behaviour when failure storage is fullDemonstrate: Choose pause or discard behaviour when failure storage is full. State the plan, limits and behaviour on failure. Use a sample event, a repeated event and a failing destination.
2.5 points maximumAsk for a demonstrationDemonstrate: Choose pause or discard behaviour when failure storage is full. State the plan, limits and behaviour on failure. Use a sample event, a repeated event and a failing destination. A scenario to run with the supplier; not a completed product test. | Not verified0–2.5 / 2.5 pts This specific check has not been verified. Missing evidence does not mean the capability is absent. Ask for a demonstration in the proposed plan. Not verified · Oct 21, 2026Source ↗ |
|---|
Depth checkStop a workflow after a configured execution timeoutDemonstrate: Stop a workflow after a configured execution timeout. State the plan, limits and behaviour on failure. Use a sample event, a repeated event and a failing destination.
2.5 points maximumAsk for a demonstrationDemonstrate: Stop a workflow after a configured execution timeout. State the plan, limits and behaviour on failure. Use a sample event, a repeated event and a failing destination. A scenario to run with the supplier; not a completed product test. | Supported2.5 / 2.5 pts Workflows can stop after a configured time limit. Timeout Workflow setting; Cloud plans enforce different maximum timeouts. Global capability evidence; local plan availability and interface language are not inferred. Official documentation · Oct 21, 2026Source ↗ |
|---|
Depth checkSuspend execution and resume from an external callbackDemonstrate: Suspend execution and resume from an external callback. State the plan, limits and behaviour on failure. Use a sample event, a repeated event and a failing destination.
2.5 points maximumAsk for a demonstrationDemonstrate: Suspend execution and resume from an external callback. State the plan, limits and behaviour on failure. Use a sample event, a repeated event and a failing destination. A scenario to run with the supplier; not a completed product test. | Supported2.5 / 2.5 pts Paused executions can resume on an external webhook. Wait node with per-execution webhook URL; authentication and timeout should be configured. Global capability evidence; local plan availability and interface language are not inferred. Official documentation · Oct 21, 2026Source ↗ |
|---|
Depth checkRestrict workflow requests to approved destinationsDemonstrate: Restrict workflow requests to approved destinations. State the plan, limits and behaviour on failure. Use a sample event, a repeated event and a failing destination.
2.5 points maximumAsk for a demonstrationDemonstrate: Restrict workflow requests to approved destinations. State the plan, limits and behaviour on failure. Use a sample event, a repeated event and a failing destination. A scenario to run with the supplier; not a completed product test. | Not verified0–2.5 / 2.5 pts This specific check has not been verified. Missing evidence does not mean the capability is absent. Ask for a demonstration in the proposed plan. Not verified · Oct 21, 2026Source ↗ |
|---|
Depth checkRoll back supported transactional changes on failureDemonstrate: Roll back supported transactional changes on failure. State the plan, limits and behaviour on failure. Use a sample event, a repeated event and a failing destination.
2.5 points maximumAsk for a demonstrationDemonstrate: Roll back supported transactional changes on failure. State the plan, limits and behaviour on failure. Use a sample event, a repeated event and a failing destination. A scenario to run with the supplier; not a completed product test. | Not verified0–2.5 / 2.5 pts This specific check has not been verified. Missing evidence does not mean the capability is absent. Ask for a demonstration in the proposed plan. Not verified · Oct 21, 2026Source ↗ |
|---|
