---
title: "Prompt Injection Detection"
method: POST
path: "/v1/prompt-injection-detect"
tags: ["security"]
---

# Prompt Injection Detection

`POST /v1/prompt-injection-detect`

## Request

- Base URL: `https://agent.pocket.network`
- URL: `https://agent.pocket.network/v1/prompt-injection-detect`
- Auth: none required

Classify agent-bound text for prompt injection: a safe, suspicious, or malicious verdict with the patterns that matched. POST /v1/prompt with {text} and get the classification as one JSON object. Deterministic per ruleset version. Pay per request in USDC; no account, no API key.

## Request body

- object: Text to classify for prompt injection.
  - `text` string, required: The agent-bound text to inspect.

## Response `200`

The upstream response, served after payment settles.

- object: For POST /v1/prompt, a JSON object with the verdict and a `matches` array of the patterns that fired. The exact shape is the service's own and is not pinned here; errors return a JSON object with an `error` field, and the GET routes return their own small JSON documents.

## Other responses

- `400`: Malformed request.
- `402`: Payment required. The x402 v2 terms are in the PAYMENT-REQUIRED header and, identically, in this body; sign the chosen option and retry with the PAYMENT-SIGNATURE header. The receipt arrives in PAYMENT-RESPONSE.
- `503`: This service is not currently available to serve.

---

[API](https://skmtc.dev/pocket/apis/pocket-agentic-portal.md) · [All operations](https://skmtc.dev/pocket/apis/pocket-agentic-portal/llms.txt) · [OpenAPI document](https://skmtc.dev/pocket/apis/pocket-agentic-portal/revisions/96d2176db014?raw)