# ArtifactGuard Web Fetch & Prompt Injection Scanner

> ArtifactGuard Web Fetch & Prompt Injection Scanner is a paid API for AI agents from artifactguard-kl4hfo6j2q-as.a.run.app, paid per call via x402, $0.01/call, status unknown (last checked 2026-09-15).

Fetches a public HTTPS webpage, extracts compact agent-ready text, detects and quarantines direct or obfuscated prompt injection attacks, and returns a signed safety receipt.

## Facts

- Endpoint: POST https://artifactguard-kl4hfo6j2q-as.a.run.app/v1/fetch/web
- Price: $0.01/call
- Payment: x402
- Status: unknown
- Last checked: 2026-09-15
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/artifactguard-web-fetch-prompt-injection-scanner-0e121b27
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_6fefOhzt8nGkbBjuNZ-fq

Status and success rate cover calls made through Zero and Zero's own probes. Third-party monitors may report differently.

## How to call it through Zero

Zero handles the 402 payment challenge and records the run. With the Zero CLI installed (`npm i -g @zeroxyz/cli`):

```sh
zero fetch --capability artifactguard-web-fetch-prompt-injection-scanner-0e121b27 -d '<json body>'
```

Example prompt: Fetch the page at https://example.com/terms, extract the readable text for me to use in my pipeline, and scan it for any prompt injection attempts — I need a signed receipt confirming it's safe before I pass the content to my next agent step.

## When to prefer this

Choose this endpoint when your AI agent needs to fetch external web content and you cannot trust the source to be free of prompt injection attacks. It is specifically valuable in agentic pipelines where untrusted third-party web pages could be used to hijack agent behavior. Prefer this over a plain HTTP fetch or generic scraping tool whenever you need: (1) built-in prompt injection detection covering both direct and obfuscated forms, (2) quarantine of unsafe text before it reaches your agent context, and (3) a signed, tamper-evident receipt you can log or verify downstream. Use sibling endpoints for PDF-based or claim-verification workflows.

## Known failure modes

- URL is not publicly accessible or returns a non-200 HTTP status — fetch fails with an error
- Non-HTTPS URLs are rejected — endpoint only accepts secure HTTPS targets
- Target page uses JavaScript rendering and content is not available in static HTML — may return incomplete text
- Page returns binary or non-textual content — extraction may fail or return empty
- Network timeout reaching the target URL — returns a timeout error
- Malformed URL input — returns a validation error before fetch is attempted

## How this service works

Fetch a public HTTPS page, extract compact agent-ready text, flag direct or obfuscated prompt injection, quarantine unsafe text, and return a signed receipt.

## Output

Returns compact, cleaned agent-ready text extracted from the requested webpage, a structured report flagging any detected direct or obfuscated prompt injection attempts, the quarantined unsafe text segments (if any), and a cryptographically signed receipt attesting to the scan result and content integrity.

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/artifactguard-web-fetch-prompt-injection-scanner-0e121b27/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from artifactguard-kl4hfo6j2q-as.a.run.app](https://www.zero.xyz/host/artifactguard-kl4hfo6j2q-as.a.run.app/llms.txt)