# canvas_run_agent

显式运行单个 Canvas Agent 并等待终态；不会自动运行下游节点或启动图片任务。专业审核用 reviewScope 明确整图或指定节点，返回 reviewCoverage 数据覆盖；首次导演规划可用 positionBeforeNodeIds 将该Agent移到制作节点前方，只改位置不加边。

Agent View of the PolicyLayer registry record for `canvas_run_agent`. HTML page: https://policylayer.com/tools/dutydeck/canvas-run-agent

## Facts

- Tool: `canvas_run_agent`
- Server: DutyDeck (`taokoplay/DutyDeck`) — https://policylayer.com/tools/dutydeck.md
- Homepage: https://github.com/taokoplay/DutyDeck
- Risk category: Execute (High risk)
- Registry record: grade F, identity unverified
- Server rate-limited: no
- Parameters: 0
- Recommended policy verdict: Rate-limited

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

```json
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "canvas_run_agent",
    "arguments": {}
  }
}
```

## Why canvas_run_agent is rated High

canvas_run_agent triggers real processes with real consequences. An agent gone sideways doesn't fire it once. It starts dozens of builds, sends mass notifications, or burns through compute before anyone looks up.

## Use case

AI agents invoke canvas_run_agent to trigger actions in DutyDeck. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call: builds kicked off, notifications sent, workflows started.

## Recommended policy (PolicyLayer)

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

```json
{
  "version": "1",
  "default": "deny",
  "tools": {
    "canvas_run_agent": {
      "limits": [
        {
          "counter": "canvas_run_agent_rate",
          "window": "minute",
          "max": 10,
          "scope": "grant"
        }
      ]
    }
  }
}
```

## Other tools on DutyDeck (246)

- `canvas_cancel_media_run` — Destructive — https://policylayer.com/tools/dutydeck/canvas-cancel-media-run.md
- `factory_cancel_experiment` — Destructive — https://policylayer.com/tools/dutydeck/factory-cancel-experiment.md
- `mcp__automation__delete_automation` — Destructive — https://policylayer.com/tools/dutydeck/mcp--automation--delete-automation.md
- `mcp__planning__delete_calendar_event` — Destructive — https://policylayer.com/tools/dutydeck/mcp--planning--delete-calendar-event.md
- `mcp__planning__delete_group` — Destructive — https://policylayer.com/tools/dutydeck/mcp--planning--delete-group.md
- `mcp__planning__delete_reminder` — Destructive — https://policylayer.com/tools/dutydeck/mcp--planning--delete-reminder.md
- `mcp__planning__delete_tag` — Destructive — https://policylayer.com/tools/dutydeck/mcp--planning--delete-tag.md
- `mcp__planning__delete_todo` — Destructive — https://policylayer.com/tools/dutydeck/mcp--planning--delete-todo.md
- `media_cancel_generation` — Destructive — https://policylayer.com/tools/dutydeck/media-cancel-generation.md
- `media_cancel_run` — Destructive — https://policylayer.com/tools/dutydeck/media-cancel-run.md
- `api_prepare_request` — Execute — https://policylayer.com/tools/dutydeck/api-prepare-request.md
- `api_run_scenario` — Execute — https://policylayer.com/tools/dutydeck/api-run-scenario.md
- `BrowserExecuteJavaScript` — Execute — https://policylayer.com/tools/dutydeck/browserexecutejavascript.md
- `BrowserNavigate` — Execute — https://policylayer.com/tools/dutydeck/browsernavigate.md
- `BrowserNewTab` — Execute — https://policylayer.com/tools/dutydeck/browsernewtab.md
- `BrowserWaitFor` — Execute — https://policylayer.com/tools/dutydeck/browserwaitfor.md
- `canvas_run_nodes` — Execute — https://policylayer.com/tools/dutydeck/canvas-run-nodes.md
- `canvas_run_workflow` — Execute — https://policylayer.com/tools/dutydeck/canvas-run-workflow.md
- `canvas_task` — Execute — https://policylayer.com/tools/dutydeck/canvas-task.md
- `factory_advance_experiment` — Execute — https://policylayer.com/tools/dutydeck/factory-advance-experiment.md
- `factory_diagnose_run` — Execute — https://policylayer.com/tools/dutydeck/factory-diagnose-run.md
- `factory_retry_review` — Execute — https://policylayer.com/tools/dutydeck/factory-retry-review.md
- `factory_run_batch` — Execute — https://policylayer.com/tools/dutydeck/factory-run-batch.md
- `factory_run_scene` — Execute — https://policylayer.com/tools/dutydeck/factory-run-scene.md
- `factory_run_step` — Execute — https://policylayer.com/tools/dutydeck/factory-run-step.md
- `mcp__automation__run_automation_now` — Execute — https://policylayer.com/tools/dutydeck/mcp--automation--run-automation-now.md
- `mcp__collaboration__stop_delegation` — Execute — https://policylayer.com/tools/dutydeck/mcp--collaboration--stop-delegation.md
- `mcp__collaboration__stop_delegations` — Execute — https://policylayer.com/tools/dutydeck/mcp--collaboration--stop-delegations.md
- `mcp__collaboration__wait_for_delegations` — Execute — https://policylayer.com/tools/dutydeck/mcp--collaboration--wait-for-delegations.md
- `media_execute_generation` — Execute — https://policylayer.com/tools/dutydeck/media-execute-generation.md
- …and 216 more: https://policylayer.com/tools/dutydeck.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=dutydeck · API: https://policylayer.com/registry/api · Policy library: https://policylayer.com/policies/dutydeck