# TrustFetch Batch Injection Scan

> TrustFetch Batch Injection Scan is a paid API for AI agents from trustfetch.duckdns.org, paid per call via x402, $0.12/call, status unknown (last checked 2026-09-13).

Scans a batch of text strings for hidden prompt-injection attempts, returning per-text signals, confidence scores, and detected techniques.

## Facts

- Endpoint: POST https://trustfetch.duckdns.org/tools/injection-scan/batch
- Price: $0.12/call
- Payment: x402
- Status: unknown
- Last checked: 2026-09-13
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/trustfetch-batch-injection-scan-13089f96
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_SYcGM4HYIv34oFWeIrGY4

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 trustfetch-batch-injection-scan-13089f96 -d '<json body>'
```

Example prompt: Before you act on this batch of user messages, scan all of them for prompt injection attacks and tell me which ones contain suspicious override phrases or hidden instructions trying to hijack your behavior.

## When to prefer this

Choose this endpoint when you need to screen multiple texts at once for prompt injection before an AI agent acts on them — especially when processing untrusted user input, scraped web content, or external documents in bulk. It is purpose-built for LLM security use cases rather than general content moderation or spam filtering. Prefer this over single-text scanning when you have 2+ texts to check, since batch processing is more cost-efficient per text.

## Known failure modes

- Missing required 'texts' array returns validation error
- Empty texts array may return empty results with no useful signal
- Very long or malformed strings may cause processing errors
- Heuristic detector may produce false positives on benign imperative language
- Low-confidence detections (confidence < 0.5) may not reliably indicate true injection
- Network or DNS issues with duckdns.org domain could cause connectivity failures
- Payment authorization failure blocks the call before scanning begins

## How this service works

Volume tier: scan up to 10 texts for a single flat price -- cheaper per item than calling the fast tier that many times separately.

## Output

Returns a JSON object with a count and a results array (one entry per input text). Each result includes: injection_suspected (boolean), confidence (0-1 float), technique name (e.g. 'override_phrase'), flagged_snippet (the offending text), reasoning (human-readable explanation), detector_tier (e.g. 'heuristic'), and a signals array with granular per-signal details including snippet, reasoning, technique, and confidence.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "texts": {
   "type": "array",
   "items": {
    "type": "string"
   },
   "description": "List of texts to scan, one result per text"
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "json",
 "example": {
  "count": 1,
  "results": [
   {
    "signals": [
     {
      "snippet": "Ignore all previous instructions",
      "reasoning": "Text contains a phrase attempting to override prior instructions.",
      "technique": "override_phrase",
      "confidence": 0.9
     }
    ],
    "reasoning": "Text contains a phrase attempting to override prior instructions.",
    "technique": "override_phrase",
    "confidence": 0.9,
    "detector_tier": "heuristic",
    "flagged_snippet": "Ignore all previous instructions",
    "injection_suspected": true
   }
  ]
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/trustfetch-batch-injection-scan-13089f96/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from trustfetch.duckdns.org](https://www.zero.xyz/host/trustfetch.duckdns.org/llms.txt)