# wasm_agent_prompt

Send a prompt to a WASM agent and get a response. Use when native Task is wrong because the workload needs sandboxed isolation — untrusted code execution, browser-side run, deterministic replay. Pair with wasm_gallery_search to find a published agent, or wasm_agent_create to scaffold a fresh one. For trusted in-process work, native Task is fine.

Agent View of the PolicyLayer registry record for `wasm_agent_prompt`. HTML page: https://policylayer.com/tools/ruflo/wasm-agent-prompt

## Facts

- Tool: `wasm_agent_prompt`
- Server: Ruflo (`ruvnet/ruflo`) — https://policylayer.com/tools/ruflo.md
- Homepage: https://github.com/ruvnet/ruflo
- 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": "wasm_agent_prompt",
    "arguments": {}
  }
}
```

## Why wasm_agent_prompt is rated High

This tool executes arbitrary prompts against WASM agents, which are described as capable of handling 'untrusted code execution' scenarios. While sandboxed, execution tools that run code (even in isolation) present high risk if an AI agent sends malicious or unintended prompts. The blast radius includes computational resource consumption, data exfiltration from the sandbox, and potential sandbox escape vectors.

From the tool's own definition: "Send a prompt to a WASM agent and get a response...untrusted code execution, browser-side run, deterministic replay. The tool explicitly mentions 'untrusted code execution' and triggers agent operations in a sandboxed environment."

## Use case

AI agents invoke wasm_agent_prompt to trigger actions in Ruflo. 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 Ruflo:

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

## Other tools on Ruflo (495)

- `agent_terminate` — Destructive — https://policylayer.com/tools/ruflo/agent-terminate.md
- `agentdb_batch` — Destructive — https://policylayer.com/tools/ruflo/agentdb-batch.md
- `agentdb_causal-edge-delete` — Destructive — https://policylayer.com/tools/ruflo/agentdb-causal-edge-delete.md
- `agentdb_causal-node-delete` — Destructive — https://policylayer.com/tools/ruflo/agentdb-causal-node-delete.md
- `agentdb_hierarchical-delete` — Destructive — https://policylayer.com/tools/ruflo/agentdb-hierarchical-delete.md
- `autopilot_reset` — Destructive — https://policylayer.com/tools/ruflo/autopilot-reset.md
- `config_reset` — Destructive — https://policylayer.com/tools/ruflo/config-reset.md
- `federation_wg_keyrotate` — Destructive — https://policylayer.com/tools/ruflo/federation-wg-keyrotate.md
- `hooks_intelligence-reset` — Destructive — https://policylayer.com/tools/ruflo/hooks-intelligence-reset.md
- `hooks_worker-cancel` — Destructive — https://policylayer.com/tools/ruflo/hooks-worker-cancel.md
- `iot_device_remove` — Destructive — https://policylayer.com/tools/ruflo/iot-device-remove.md
- `iot_device_unpair` — Destructive — https://policylayer.com/tools/ruflo/iot-device-unpair.md
- `iot_firmware_rollout_rollback` — Destructive — https://policylayer.com/tools/ruflo/iot-firmware-rollout-rollback.md
- `iot_fleet_delete` — Destructive — https://policylayer.com/tools/ruflo/iot-fleet-delete.md
- `iot_fleet_remove_device` — Destructive — https://policylayer.com/tools/ruflo/iot-fleet-remove-device.md
- `managed_agent_terminate` — Destructive — https://policylayer.com/tools/ruflo/managed-agent-terminate.md
- `memory_delete` — Destructive — https://policylayer.com/tools/ruflo/memory-delete.md
- `policy_revoke` — Destructive — https://policylayer.com/tools/ruflo/policy-revoke.md
- `ruvector_delete` — Destructive — https://policylayer.com/tools/ruflo/ruvector-delete.md
- `session_delete` — Destructive — https://policylayer.com/tools/ruflo/session-delete.md
- `system_reset` — Destructive — https://policylayer.com/tools/ruflo/system-reset.md
- `task_cancel` — Destructive — https://policylayer.com/tools/ruflo/task-cancel.md
- `wasm_agent_reset` — Destructive — https://policylayer.com/tools/ruflo/wasm-agent-reset.md
- `wasm_gallery_remove_custom` — Destructive — https://policylayer.com/tools/ruflo/wasm-gallery-remove-custom.md
- `workflow_cancel` — Destructive — https://policylayer.com/tools/ruflo/workflow-cancel.md
- `workflow_delete` — Destructive — https://policylayer.com/tools/ruflo/workflow-delete.md
- `agent_pool` — Execute — https://policylayer.com/tools/ruflo/agent-pool.md
- `agent_spawn` — Execute — https://policylayer.com/tools/ruflo/agent-spawn.md
- `agentdb_consolidate` — Execute — https://policylayer.com/tools/ruflo/agentdb-consolidate.md
- `agentdb_route` — Execute — https://policylayer.com/tools/ruflo/agentdb-route.md
- …and 465 more: https://policylayer.com/tools/ruflo.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=ruflo · API: https://policylayer.com/registry/api · Policy library: https://policylayer.com/policies/ruflo