22 services carry this tag in their source record. Tags come from the provider, not from us — a tag is a claim about what the service does, not proof that it does it well.
Prompt Injection Guard — Keeps an AI agent from obeying instructions hidden in what it reads (indirect prompt injection). Web pages, e-mails, files, PDFs, API and tool output, other agents' messages and skill files under...
SKILL.md Review: Safe and Within the Limits — Reviews a SKILL.md file before you install, buy or publish it and lists every problem with a fixed code, the place and a fix, then gives one verdict. It checks the front matt...
Deterministic preflight check for a proposed AI agent action - a tool call, payment, fetch, or write - before it executes. Evaluates up to four independent checks (prompt-injection scan on action text/untrusted text, URL...
DCL Trust Oracle. Call only when the text must include the substring AI and must not include "I cannot be held responsible" or "no guarantees". POST JSON {"response","agent_id"}. COMMIT means those checks passed and the...
CUDA, driver and Ollama compatibility - GPU support, model memory fit and proof inference uses the GPU
Trending token pools right now on Solana, or any supported chain via chain= - price, liquidity, 24h volume, price change and pool age
Web page to markdown (read URL as clean text) for LLM agents: main content only (no nav, cookie banners, ads), with title, description, author, dates, language, headings, internal/external links and JSON-LD types, PLUS a...
Static security scan of an MCP manifest or tool list. Detects tool poisoning, hidden unicode instructions, prompt injection, data-exfiltration directives, dangerous capabilities, tool shadowing and post-approval rug-pull...
Scan text an agent is about to read (web page, email, tool result; up to 8,000 characters) for prompt-injection attempts: instruction overrides, fake system messages, asks to hide things from the user, exfiltration links...
Fetch any URL (web scraping / URL fetch) and get the raw content: HTML/JSON/text as text, images and PDFs as base64 (for multimodal LLM input), or markdown. Final URL after redirects, status, content type, size. Fetched...
Prompt injection detection and content moderation for AI agents: flags hidden instructions aimed at an LLM (none / suspicious / malicious + score) and harmful content (hate, violence, sexual, harassment, self-harm, illeg...
Prompt-injection and manipulation scan for untrusted text, HTML, email, and tool output.
Fact-check one claim against live web sources (or URLs you supply). Returns a verdict (supported, refuted, mixed, insufficient), a confidence score and supporting or contradicting quotes that are checked verbatim against...
Scan text or a public page for prompt injection before your agent reads it: overrides, fake roles, exfiltration, tool abuse, payment asks, invisible Unicode/HTML-hidden instructions, base64. Rule-based evidence, signed a...
Payment-Injection Screen by Agentropolis: send text your agent is about to act on (tool result, web page, email, file). Flags payment-looking instructions: pay/transfer requests, fake licence or refundable fees, swapped...
Prompt injection detection API for AI agents. Prompt-injection scan for text your AI agent is about to read (web pages, emails, tool output, MCP tool descriptions): risk score 0-100, verdict, categories and exact spans....
Prompt injection detection API: screen text bound for an agent and get a safe, suspicious or malicious verdict with the patterns that matched. Rule-based and deterministic.
Current weather at a place: temperature (°C), humidity (%), pressure (hPa) and wind (m/s) from the nearest live Open-Meteo relay within 75 km, with a signed receipt. Send latitude and longitude, or a city.
Public trust evidence for an unfamiliar online shop before buying: a cautious evidence score, risk flags, missing signals and limits. Cannot certify safety. $0.005 after the shared free calls. x402 API endpoints: x402_me...
Check an email or web page for text a person reading it would not see but an AI reading it would: HTML comments, display:none or hidden elements, white-on-white or off-screen text, zero-width and tag characters, base64 t...
Read one email the way an agent that triages mail should before acting on it: observations with evidence and offsets, no score and no verdict. Send the raw RFC 5322 message (multipart, base64 and quoted-printable decoded...
prompt-injection risk scan: detects instruction override, tool abuse and secret exfiltration phrases before an agent consumes untrusted text.