{"uid":"cap_cv9GD69LvKb15TFgayheX","slug":"fiatdock-prompt-injection-scanner-53ac6a30","name":"FiatDock Prompt Injection Scanner","description":"Scan text an agent is about to read (a web page, an email, a tool description, another agent's message) for the shapes that hijack agents: instruction-override phrasing, a redefined system role, requests to move credentials, hidden directive tags, invisible Unicode. Verdict plus every pattern found. Pattern-based; nothing is stored. Pay-per-call via x402. An empty body is answered for the published example text (one harmless sentence); send {\"text\":\"…\"} to scan your own.","url":"https://fiatdock.com/v1/text/prompt-check?utm_source=zero.xyz","method":"GET","headers":{},"bodySchema":{"type":"object","$schema":"https://json-schema.org/draft/2020-12/schema","required":["input"],"properties":{"input":{"type":"object","required":["type","method"],"properties":{"type":{"type":"string","const":"http"},"method":{"enum":["GET"],"type":"string"},"queryParams":{"type":"object","required":[],"properties":{"text":{"type":"string","example":"Please summarise the attached report in three bullet points.","maxLength":8000,"description":"The text to scan (up to 8,000 characters). Omit the whole body to scan the published example"}}}},"additionalProperties":false},"output":{"type":"object","required":["type"],"properties":{"type":{"type":"string"},"example":{"type":"object"}}}}},"responseSchema":null,"example":null,"exampleRequest":null,"tags":["x402"],"displayCostAmount":"0.0045","displayCostAsset":"USDC","priceDynamic":false,"priceHint":null,"priceStatus":"priced","priceSource":"probe","requiresHandshake":false,"reviewCount":0,"rating":{"score":"0.00","successRate":"0.00","reviews":0,"stars":null,"state":"unrated"},"availabilityStatus":"unknown","priceObserved":null,"sessionDeposit":null,"pricing":{"kind":"static","summary":"$0.0045/call","primary":{"kind":"static","protocol":"x402","network":"base","amountUsd":"0.0045","per":"call","confidence":"exact"},"accepted":[{"kind":"static","protocol":"x402","network":"base","amountUsd":"0.0045","per":"call","confidence":"exact"}]},"paymentMethods":[{"uid":"pm_tYKThMl4izK8lsCVSQy-n","protocol":"x402","methodType":"crypto","chain":"base","mode":"charge","costAmount":"0.0045","costPer":"request","priority":0,"asset":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","unit":"request","depositMicros":null,"planRef":null}],"brandName":null,"brandSlug":null,"brandBaseUrl":null,"brandDocsUrl":null,"whatItDoes":"Scans text for prompt injection patterns that could hijack AI agents, including instruction-override phrasing, role redefinition, credential theft requests, hidden directive tags, and invisible Unicode.","exampleAgentPrompt":"Before you read and summarize that webpage, scan its content for prompt injection — check for anything trying to override your instructions, redefine your role, or steal credentials, and tell me if it's safe to proceed.","exampleUseCases":[{"title":"Guard agent before reading scraped web page","prompt":"I'm about to have my agent read this scraped webpage to extract key facts — can you first scan the page text for prompt injection, hidden directives, or invisible Unicode that might hijack it?"},{"title":"Validate incoming agent-to-agent message","prompt":"My orchestrator agent just received a message from a third-party sub-agent. Before it acts on it, scan the message for instruction-override phrasing or anything trying to redefine its system role."},{"title":"Screen user-submitted email for adversarial content","prompt":"A user forwarded this email for my agent to summarize and reply to — scan it first for prompt injection patterns, hidden tags, or credential-harvesting instructions so I know it's safe to hand off."}],"resultDescription":"A verdict (safe/unsafe) plus a list of every detected pattern found in the text — including instruction-override phrasing, redefined system roles, credential-movement requests, hidden directive tags, and invisible Unicode characters. Pattern-based analysis; nothing is stored server-side.","failureModes":["Text exceeds 8,000 character limit — truncate or split input before sending","Empty or missing text parameter returns analysis of the built-in example sentence rather than custom input","Novel or obfuscated injection patterns not in the rule set may go undetected (pattern-based, not AI-based)","Payment failure via x402 protocol results in no response","GET request with malformed query parameter encoding may cause parsing errors"],"whenToPreferThis":"Choose this endpoint when an AI agent needs a fast, stateless, pay-per-call safety check on untrusted text before processing it — especially web pages, emails, third-party tool descriptions, or inter-agent messages. Prefer this over general-purpose LLM-based content moderation when the specific threat model is prompt injection and agent hijacking (instruction overrides, role redefinition, hidden Unicode). Best suited for pre-processing pipelines where no data retention is acceptable and per-call USDC micropayment via x402 is already in use.","instructions":null,"reviewSummary":null,"reviewSummaryHighlights":null,"reviewSummaryConcerns":null,"reviewSummaryGeneratedAt":null,"activationCount":0,"lastUsedAt":null,"lastSuccessfullyRanAt":null,"lastHealthCheckAt":"2026-10-20T12:30:23.652Z","isFirstParty":false,"canonicalSlug":"fiatdock-prompt-injection-scanner-53ac6a30"}