# Text Mojibake Repair

> Text Mojibake Repair is a paid API for AI agents from www.tradepilotusa.com, paid per call via x402, $0.01/call, status unknown (last checked 2026-10-01).

Repairs UTF-8-to-Latin-1 mojibake corruption in text by pattern-matching known garbled character sequences back to their intended Unicode characters

## Facts

- Endpoint: POST https://www.tradepilotusa.com/api/agent-commerce/v1/services/text_mojibake_repair/execute?utm_source=zero.xyz
- Price: $0.01/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-01
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/text-mojibake-repair-e86f40d6
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_az0MJKP6ISBaOw5kKDcui

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 text-mojibake-repair-e86f40d6 -d '<json body>'
```

Example prompt: This customer review has garbled characters everywhere — curly quotes are showing as “ and ’ and em dashes look like â€

## When to prefer this

Use this endpoint when you have text that was clearly corrupted by the classic UTF-8-bytes-interpreted-as-Latin-1 mistake, showing symptoms like “ for curly quotes, â€

## Known failure modes

- Text that is not mojibake-corrupted is returned unchanged — no false positives but no detection confirmation either
- Uncommon or non-standard corruption patterns not in the heuristic list will not be repaired
- Text exceeding 48,000 characters will be rejected with a validation error
- Non-UTF-8/Latin-1 encoding errors (e.g. Windows-1252 specific sequences) may not be fully addressed
- Payment failure or insufficient USDC balance will prevent the call from completing

## How this service works

Repair text corrupted by the classic UTF-8-bytes-read-as-Latin-1 mistake (curly quotes, em dashes, common emoji, stray BOM appearing as garbled multi-character sequences). Pattern-matches known corruption sequences; this is a heuristic repair, not full charset detection, and leaves correctly-encoded text unchanged.

## Output

Returns the input text with known UTF-8-read-as-Latin-1 corruption sequences replaced by their correct Unicode characters. Correctly-encoded text is returned unchanged. The repair is heuristic and pattern-based, not a full charset detection solution.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "text": {
   "type": "string",
   "maxLength": 48000
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/text-mojibake-repair-e86f40d6/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from www.tradepilotusa.com](https://www.zero.xyz/host/www.tradepilotusa.com/llms.txt)