# Verdict L402 – Web Guard (Hidden Instruction Detector)

> Verdict L402 – Web Guard (Hidden Instruction Detector) is a paid API for AI agents from verdict.kechlab.com, paid per call via x402, $0.0214/call, status unknown (last checked 2026-10-21).

Scans an HTML page for hidden text invisible to humans (HTML comments, CSS-hidden elements, hidden attributes, Unicode-encoded content) and determines whether those hidden instructions are targeting an AI agent.

## Facts

- Endpoint: POST https://verdict.kechlab.com/v1/web-guard?utm_source=zero.xyz
- Price: $0.0214/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-21
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/verdict-l402-web-guard-hidden-instruction-detector-14b13966
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_kJcUPRS52A0aQ4HUUipcv

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 verdict-l402-web-guard-hidden-instruction-detector-14b13966 -d '<json body>'
```

Example prompt: Before you summarize that webpage, scan its raw HTML for any hidden instructions targeting you — things like CSS-invisible text, HTML comments with orders, or Unicode-encoded commands — and tell me whether the page is trying to manipulate you.

## When to prefer this

Use this endpoint when an AI agent is about to read, summarize, or act on a web page fetched from an untrusted source and you need to detect prompt injection attacks hidden in HTML that are invisible to human readers. It is specifically designed for the web-browsing agent attack surface (HTML comments, CSS hidden text, Unicode steganography) rather than general content moderation or visible text analysis.

## Known failure modes

- HTML exceeds 400 KB limit — request rejected
- Malformed or non-HTML input — parsing failure
- No hidden content found — clean verdict returned (not an error)
- Payment not included or insufficient USDC — 402 Payment Required
- Ambiguous or borderline hidden text — verdict may have lower confidence

## How this service works

For agents that read web pages: finds what a human cannot see but an agent reads (HTML comments, CSS-hidden text, hidden attributes, invisible Unicode-encoded text) and tells whether those hidden instructions target an AI. Extraction by code, reading by the decision engine on the hidden text only. R

## Output

Returns an analysis of hidden content found in the submitted HTML — extracted invisible text (from HTML comments, CSS-hidden elements, hidden attributes, and Unicode-encoded sequences) — along with a verdict on whether those hidden instructions appear to be targeting an AI agent.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "html": {
   "type": "string",
   "description": "the HTML of the page (400 KB maximum)"
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/verdict-l402-web-guard-hidden-instruction-detector-14b13966/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from verdict.kechlab.com](https://www.zero.xyz/host/verdict.kechlab.com/llms.txt)