PromptShield

active

Pay-per-request prompt-injection screener and HTML cleaner for AI agents. USDC on Base via x402 v2. No API keys, no signup.

Settled via Coinbase.

Transactions · 30d
163
Volume · 30d
$0.475
Unique buyers · 30d
14
Uptime · 30d
100.0%
Latency p50
63ms
Reported calls · 30d
20

Endpoints (4 live)

  • POST /v1/screen — Screen a support ticket, inbound email, scraped page or tool output for hidden instructions before your agent reads it. Returns risk (low|medium|high), findings, and sanitized text. $0.002 USDC on Base. (0.002 USDC on Base)
  • POST /v1/clean — Turn a docs page, email or scraped HTML into clean LLM-ready text: strips scripts, styles, hidden elements and zero-width characters; returns text plus a token estimate. $0.002 USDC on Base. (0.002 USDC on Base)
  • GET /v1/clean — Turn a docs page, email or scraped HTML into clean LLM-ready text: strips scripts, styles, hidden elements and zero-width characters; returns text plus a token estimate. $0.002 USDC on Base. GET works with no input (cleans a built-in sample) or ?content=<url-encoded html>; prefer POST with JSON body for real pages. (0.002 USDC on Base)
  • GET /v1/screen (0.002 USDC on Base)

First seen · last seen · last active