# test_run_cloudflow_flow

Manage CloudFlow. Runs a flow once as a test, and accepts an unpublished (draft) flow — unlike actions/trigger, which requires the flow to be published. Use this to verify a newly authored or edited flow before publishing it. Execution is identical to a production run: the same graph, the same bound connections, the same credentials, and the same approval behaviour. An approval-gated node still parks the run and its side effect still waits for a real approval — there is no request that skips one. The run is recorded as a test, so it does not appear in run history, does not count towards dashboard statistics, does not become a dashboard widget's data source, and does not consume a scheduled-run budget. It does hold the flow's concurrency lock, so a 409 is returned while the flow is already running. The flow's first node must be a webhook, scheduled, or manual trigger. Before dispatching, the flow is validated with the same checks publish applies — a draft has never been through them. A flow that fails returns 422 listing every offending node at once, so all of them can be fixed in one pass. Poll the Location URL to follow the run and read what each node produced. Pass ?dryRun=true to validate the flow without starting a run. Generated codeNode code may fail silently or error at run time despite passing validation. With dryRun this call shows only that the flow is well-formed; without dryRun it executes real actions even on drafts. A completed run's per-node output (get_cloudflow_flow_run) shows whether it works. The current generated MCP schema exposes no arbitrary trigger payload field; the API uses an empty payload when omitted. Idempotency-Key is required even with dryRun. Same key/request replays within 24 hours; different request fails with 422, an in-progress match with 409. Dry-runs validate existing fingerprints without storing a replay. The API fingerprints MCP tracking parameters; a changed client/server version can cause a same-key conflict. Keep the original request context and key. If an HTTP failure has only generic text, do not infer a status or retry automatically.

Agent View of the PolicyLayer registry record for `test_run_cloudflow_flow`. HTML page: https://policylayer.com/tools/doit/test-run-cloudflow-flow

## Facts

- Tool: `test_run_cloudflow_flow`
- Server: Doit (`@doitintl/doit-mcp-server`) — https://policylayer.com/tools/doit.md
- Install: `npx -y @doitintl/doit-mcp-server`
- Homepage: https://www.npmjs.com/package/@doitintl/doit-mcp-server
- Risk category: Read (Low risk)
- Registry record: grade F, identity unverified
- Server rate-limited: no
- Parameters: 4 (2 required)
- Recommended policy verdict: Allowed

## Parameters

| Parameter | Type | Required | Description |
| --- | --- | --- | --- |
| `dryRun` | boolean | no |  |
| `flowId` | string | yes |  |
| `Idempotency-Key` | string | yes |  |
| `customerContext` | string | no | Scope the request to a specific customer by ID. Required for DoiT employees (whose token isn't tied to a single customer); omit for direct customer users. |

Parameters from the server's own tool schema.

## Example call (MCP tools/call, JSON-RPC 2.0)

```json
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "test_run_cloudflow_flow",
    "arguments": {
      "flowId": "<flowId>",
      "Idempotency-Key": "<Idempotency-Key>"
    }
  }
}
```

## Why test_run_cloudflow_flow is rated Low

Even though test_run_cloudflow_flow only reads data, uncontrolled read access leaks sensitive information and racks up API costs: an agent caught in a retry loop can make thousands of calls a minute without anyone noticing.

Risk signals: Bulk/mass operation — affects multiple targets

## Use case

AI agents call test_run_cloudflow_flow to retrieve information from Doit without modifying anything. It is typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.

## Recommended policy (PolicyLayer)

Verdict: **Allowed**. Enforced by the PolicyLayer MCP gateway (https://policylayer.com/mcp-gateway) before a call reaches Doit:

```json
{
  "version": "1",
  "default": "deny",
  "tools": {
    "test_run_cloudflow_flow": {}
  }
}
```

## Other tools on Doit (237)

- `cancel_async_operation` — Destructive — https://policylayer.com/tools/doit/cancel-async-operation.md
- `cancel_invite` — Destructive — https://policylayer.com/tools/doit/cancel-invite.md
- `confirm_action` — Destructive — https://policylayer.com/tools/doit/confirm-action.md
- `delete_account_role` — Destructive — https://policylayer.com/tools/doit/delete-account-role.md
- `delete_alert` — Destructive — https://policylayer.com/tools/doit/delete-alert.md
- `delete_allocation` — Destructive — https://policylayer.com/tools/doit/delete-allocation.md
- `delete_annotation` — Destructive — https://policylayer.com/tools/doit/delete-annotation.md
- `delete_ava_conversation` — Destructive — https://policylayer.com/tools/doit/delete-ava-conversation.md
- `delete_budget` — Destructive — https://policylayer.com/tools/doit/delete-budget.md
- `delete_cloudflow_connection` — Destructive — https://policylayer.com/tools/doit/delete-cloudflow-connection.md
- `delete_custom_theme` — Destructive — https://policylayer.com/tools/doit/delete-custom-theme.md
- `delete_customer_geographic_access_scope` — Destructive — https://policylayer.com/tools/doit/delete-customer-geographic-access-scope.md
- `delete_customer_group` — Destructive — https://policylayer.com/tools/doit/delete-customer-group.md
- `delete_datahub_dataset` — Destructive — https://policylayer.com/tools/doit/delete-datahub-dataset.md
- `delete_datahub_datasets` — Destructive — https://policylayer.com/tools/doit/delete-datahub-datasets.md
- `delete_datahub_events_by_filter` — Destructive — https://policylayer.com/tools/doit/delete-datahub-events-by-filter.md
- `delete_folder` — Destructive — https://policylayer.com/tools/doit/delete-folder.md
- `delete_geographic_access_custom_region` — Destructive — https://policylayer.com/tools/doit/delete-geographic-access-custom-region.md
- `delete_insight_result` — Destructive — https://policylayer.com/tools/doit/delete-insight-result.md
- `delete_insight_results` — Destructive — https://policylayer.com/tools/doit/delete-insight-results.md
- `delete_label` — Destructive — https://policylayer.com/tools/doit/delete-label.md
- `delete_report` — Destructive — https://policylayer.com/tools/doit/delete-report.md
- `delete_role` — Destructive — https://policylayer.com/tools/doit/delete-role.md
- `delete_service_account` — Destructive — https://policylayer.com/tools/doit/delete-service-account.md
- `delete_service_account_token` — Destructive — https://policylayer.com/tools/doit/delete-service-account-token.md
- `delete_user` — Destructive — https://policylayer.com/tools/doit/delete-user.md
- `delete_user_geographic_access_scope` — Destructive — https://policylayer.com/tools/doit/delete-user-geographic-access-scope.md
- `id_of_asset` — Destructive — https://policylayer.com/tools/doit/id-of-asset.md
- `remove_ticket_tags` — Destructive — https://policylayer.com/tools/doit/remove-ticket-tags.md
- `async_run_inline` — Execute — https://policylayer.com/tools/doit/async-run-inline.md
- …and 207 more: https://policylayer.com/tools/doit.md

## For agents

This record is a snapshot. Live verdicts and the full registry:

- Check every server in your MCP config at once: `npx -y policylayer stack`
- Vet a server before you add it: install the mcp-precheck skill — `npx skills add https://policylayer.com` (skill text: https://policylayer.com/skill.md)
- Query the registry over MCP: endpoint `https://api.policylayer.com/mcp` — tools `check_mcp_server`, `check_mcp_stack`, `check_tool`, `search_registry`, `get_change_events`

---

Source: the PolicyLayer MCP registry — one continuously verified record per MCP server. Full record: https://policylayer.com/registry?q=doit · API: https://policylayer.com/registry/api · Policy library: https://policylayer.com/policies/doit