# TrustFetch

> Safety and content tools for AI agents. Scan text for hidden prompt-injection attempts, or fetch and clean a webpage -- with built-in injection screening -- before your agent acts on what it reads.

## Tools

- POST /tools/injection-scan ($0.02) -- Fast tier: scan text for prompt-injection techniques before your agent processes it. Heuristics first, LLM only on ambiguous cases.
- POST /tools/injection-scan/deep ($0.10) -- Deep tier: always escalates to an LLM verdict, skipping the fast tier's confidence-gate shortcuts. For high-stakes untrusted input (about to execute code or approve a transaction).
- POST /tools/injection-scan/batch ($0.12) -- Volume tier: scan up to 10 texts for a single flat price -- cheaper per item than calling the fast tier that many times separately.
- POST /tools/fetch-clean ($0.015) -- Fetch a URL and extract clean, boilerplate-free Markdown from it -- SSRF-guarded, size-capped, redirect-safe.
- POST /tools/fetch-clean/batch ($0.10) -- Volume tier: fetch and clean up to 5 URLs for a single flat price -- cheaper per item than calling fetch-clean that many times separately. A bad URL in the batch doesn't fail the whole call -- that item just comes back with an error, the rest still succeed.
- POST /tools/safe-fetch ($0.03) -- Flagship bundle: fetch a URL, extract clean Markdown, and injection-scan the result in one priced call -- cheaper than calling fetch-clean and injection-scan separately.

All tools are paid via x402 (USDC on Base & Solana, Coinbase CDP facilitator).

## Also available

- MCP (Streamable HTTP): /mcp/
- OpenAPI schema: /openapi.json
- Interactive docs: /docs
- Health check (free): /healthz