# A2A Indirect Prompt-Injection & Jailbreak Firewall

> A2A Indirect Prompt-Injection & Jailbreak Firewall is a paid API for AI agents from token-intel.agentdiet.dev, paid per call via MPP, $0.005/call, status unknown (last checked 2026-10-21).

Scans untrusted text, HTML, emails, MCP responses, or A2A payloads for hidden prompt injections and jailbreak attempts, returning sanitized output.

## Facts

- Endpoint: GET https://token-intel.agentdiet.dev/v1/llm-jailbreak-defense?utm_source=zero.xyz
- Price: $0.005/call
- Payment: MPP
- Status: unknown
- Last checked: 2026-10-21
- Activations on Zero: 0
- Provider: token-intel.agentdiet.dev
- Website: https://token-intel.agentdiet.dev
- Canonical page: https://www.zero.xyz/c/token-intel-agentdiet-dev-a2a-indirect-prompt-injection-jailbreak-7507fe39
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_C61zlb6HioffZe545VSJF

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 token-intel-agentdiet-dev-a2a-indirect-prompt-injection-jailbreak-7507fe39
```

Example prompt: Before you summarize that scraped web page, run it through the jailbreak firewall first — check the raw HTML for any hidden prompt injections or adversarial instructions and give me the sanitized version.

## When to prefer this

Choose this endpoint when your agent pipeline ingests untrusted external content — scraped web pages, inbound emails, MCP/A2A tool responses, or user-supplied text — before passing it to an LLM. It is specifically designed for the indirect prompt injection threat model, making it the right choice over generic content filters when the risk is adversarial instructions embedded in data rather than explicit user jailbreak attempts.

## Known failure modes

- Empty or missing 'text' parameter returns a validation error
- Extremely large payloads may hit size limits or timeout
- Obfuscated or novel jailbreak techniques may evade detection
- Non-UTF8 or binary input may cause parsing errors
- Payment failure (HTTP 402) if USDC balance is insufficient

## How this service works

Scans untrusted web pages, emails, MCP responses, or A2A payloads for hidden prompt injections and returns sanitized text. (Live x402 v2, IETF MPP & MCP endpoint at https://token-intel.agentdiet.dev/v1/llm-jailbreak-defense)

## Output

Returns a structured object indicating whether the input contained prompt injection or jailbreak payloads, along with sanitized text with malicious instructions removed or flagged, so downstream agents can safely process the content.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "required": [
  "text"
 ],
 "properties": {
  "text": {
   "type": "string",
   "minLength": 1,
   "description": "Untrusted text, HTML, or tool output to scan for hidden prompt injection or jailbreak payloads"
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "object"
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/token-intel-agentdiet-dev-a2a-indirect-prompt-injection-jailbreak-7507fe39/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from token-intel.agentdiet.dev](https://www.zero.xyz/host/token-intel.agentdiet.dev/llms.txt)