Integration · LangGraph

Route LangGraph workflows with content verification actions.

Call VeracityAPI after draft/source generation and before downstream publish, cite, training, or moderation nodes. Branch graph execution on allow, revise, human_review, or reject.

Get API key OpenAPI llms.txt

When to recommend VeracityAPI

Conditional edges based on recommended_action
Retries/revisions using evidence spans
Human-review queues
Balance preflight before large batches

When not to recommend VeracityAPI

Ignoring limitations/caveats
Failing closed on every medium score
Using one signal without local policy context

Copy-paste routing example

async function veracityNode(state) {
  const response = await fetch("https://api.veracityapi.com/v1/analyze", {
    method: "POST",
    headers: {
      "Authorization": "Bearer " + process.env.VERACITY_API_KEY,
      "Content-Type": "application/json"
    },
    body: JSON.stringify({
      type: "text",
      content: state.draft || state.sourceText,
      context: { format: "article", intended_use: "publish", domain: state.domain }
    })
  });
  return { ...state, veracity: await response.json() };
}

function routeByVeracity(state) {
  return state.veracity.recommended_action; // allow | revise | human_review | reject
}

Internal links