Verified Workflows as a LangChain tool

Install

Install the integration and add it to your Langchain workspace.

pip install verifiedworkflows-langchain

[project]
dependencies = [
  "verifiedworkflows>=1.0",
  "verifiedworkflows-langchain>=0.3",
  "langchain>=0.2"
]

Submit a task

Instantiate the client, pass it to a tool factory, and hand the resulting tool list to any agent. The tool returns a structured verdict your chain can branch on.

from verifiedworkflows import VerifiedWorkflows

client = VerifiedWorkflows(api_key="vw_live_xxx")

task = client.tasks.create(
    content="Body of the generated support reply",
    content_type="text",
    callback_url="https://your-app.example.com/webhook/vw",
)

print(task["id"])  # tsk_live_f8a29

import { VerifiedWorkflows } from "verifiedworkflows";

const client = new VerifiedWorkflows({ apiKey: "vw_live_xxx" });

const task = await client.tasks.create({
  content: "Body of the generated support reply",
  contentType: "text",
  callbackUrl: "https://your-app.example.com/webhook/vw",
});

console.log(task.id);  // tsk_live_f8a29

# Submit any AI output for human review
curl -X POST https://api.verifiedworkflows.com/v1/tasks \
  -H "Authorization: Bearer vw_live_xxx" \
  -H "Content-Type: application/json" \
  -d '{
    "content": "Body of the generated support reply",
    "content_type": "text",
    "callback_url": "https://your-app.example.com/webhook/vw"
  }'

Operations

• submit_review tool — queue content for human review; agent receives a structured verdict.
• get_review_status tool — poll a previously submitted task; useful for long-running chains.
• wait_for_review tool — block the chain until the review completes (synchronous mode).
• review_history tool — retrieve a paginated list of past tasks for memory or analytics.

Setup

Install the package and set the VW_API_KEY environment variable. The client picks it up automatically.

Async flows

For long-running agents, submit a task with a callback_url and let your chain continue. React to the webhook in a separate LangServe endpoint when the result lands.