Paid services on Base matching “detect”

Detects the likely real file format from filename, file creation time, file size, and the first 200 bytes of the file. Identifies common formats such as PDF, DOCX, XLSX, CSV, JPG/JPEG, PNG, WEBP, HEIC, TXT, HTML, MP3, WAV, MP4, AVI, MPEG, and WEBM, as well as about 15,000 other file types.

Call when qualifying a short company website list. Processes 2-5 URL slots for one fixed payment and returns evidenced profiles in input order, with per-URL failures. Batch deadline 18 seconds; each target at most 8.5 seconds. Settles only if at least half are delivered and no target was left unstarted by the deadline; otherwise returns all items uncharged. No contact values or outreach.

Purpose-bound Adobe Fonts (Typekit) observation (fonts) using the existing fresh Stack-Monitor. Returns stack_adobe_fonts_decision, direct versus inferred status, evidence, optional version difference against a caller-owned baseline and reusable snapshot. Missing fingerprints do not prove absence. No background scheduling, CVE, compatibility or installed-extension inventory. Existing outbound safe

Payment security firewall scan for an OUTGOING x402 payment: PII/secret leak detection in payment metadata, nonce replay detection, overpayment detection, prompt-injection-triggered payment analysis (fast + deep tiers), canonical-USDC asset verification, merchant pinning, velocity/spend caps, and counterparty reputation cross-check. Returns allow/flag/block with per-check reasons and an Ed25519-si

Scores JSONL rows using fixed rules and adds _quality. Detects some PII and bad-word patterns; optional redaction and score filtering. Detection is incomplete and can produce false positives. No AI.

Overwing guardrail: check text before you send, post, or act on it. Content moderation, PII detection, confidential leak detection, toxicity, self-harm and severity in ~150 ms. Returns a verdict plus one recommended_action (block / redact / review / allow) with calibrated confidence. Pass context (recipient, channel, owns_contact_info) with the outbound-message set. $0.002 USDC, no account.

6 email validation checks in one call — syntax (RFC 5322), MX record verification, disposable detection (5,000+ domains), free provider identification, role-based address detection (admin@, info@), and typo suggestion. Returns a 0-1 confidence score with detailed per-check results. POST method protects email addresses from access logs. No API keys, no multiple service integrations needed. Accepts

Audio transcription and speech to text: transcribe a voice note, meeting, podcast or call recording into text with timestamps, in about 100 languages with automatic language detection. Runs OpenAI Whisper (small by default, base for speed) on agentsvc.io itself, the audio is deleted after the call. Send audio_url or audio_base64 (mp3, mp4, m4a, wav, ogg, webm, flac), up to 10 minutes and 24 MB per

Trace the redirect chain of a URL to its final destination without downloading page bodies: every hop with its status code, time and Location header, the final URL, whether the domain changed, and loop detection. Useful before trusting a short link or an ad link. Query: ?url=https://bit.ly/…

Cross-platform arbitrage detection between Polymarket and Kalshi: matched markets, price spreads, fee-adjusted EV. Coverage: liquid Polymarket markets in politics/geopolitics/elections/tech/economy/finance; low-volume long-tail, sports, culture, and weather are not included.

Use when an agent must decide who an IP belongs to before blocking or trusting it: one IPv4 or IPv6 address against the two authoritative, keyless public sources - RDAP registry data (holder, country, date, abuse contact) via the ARIN-to-RIR chain, and the origin ASN from Team Cymru's DNS service. No city-level geolocation and no VPN/proxy detection: that needs a licensed database this route delib

Recurring wallet monitor for crypto agents. Check balances and new USDC/ERC-20 transfers since the prior cursor, then return a ready next request. Designed to be called about every 60 seconds for deposit detection, payment monitoring and treasury alerts.

Address security check — detect malicious wallets, known phishing addresses, scam contracts, and blacklisted entities across all EVM chains. Chain-independent lookup against multi-chain security databases. No API keys needed. Accepts USDC payments on Base and Solana

URL safety verdict: follows the redirect chain, detects homograph and punycode hostname tricks, and cross-references the destination host against the URLhaus threat feed. Returns safe, suspicious or malicious with reasons. Single-URL scope; domain-wide grading is /sitegrade and registration facts are /domain-intel.

Free tier: 25 calls/day per IP with header X-Free-Call: 1 (or ?free=1), then pay per call. Rule-based prompt-injection screen for untrusted text such as scraped web pages, emails and tool output. JSON body {text} (max 50,000 chars). Detects instruction-override phrasing, role-spoofing tags, data-leak links and hidden Unicode in 6 languages. Returns verdict safe, suspicious or malicious, a 0-100 sc

Token honeypot / security risk check before buy or approve. Input ERC20 contract + chain. Returns verdict avoid|caution|ok, is_honeypot, sell/buy tax, blacklist, owner powers. Use for honeypot detection, scam token screening, rug check, token security on Base Ethereum BSC Arbitrum Optimism Polygon.

For model cascades: is the small model's answer usable, or should the larger model be paid? Detects refusal, evasion, off-topic answers, contradiction and incompleteness; decides accept, review or escalate. Does not fact-check without a source: provide the source to check substance.

Deterministic prompt-injection firewall and honeypot scrubber for AI agents. Sanitizes untrusted text and blocks detected wallet-draining commands before context ingestion.

Crypto market regime classification (choppy / uptrend / downtrend / high-volatility) for SOL, BTC, ETH, XRP, DOGE, LTC, ADA, or XLM, plus a grid-trading suitability verdict. Computed live from 1h klines, a 30-day volatility baseline, and EMA trend detection — the same classifier that steered our production grid bot before we retired it 2026-08-16. For agents deciding whether a pair is grid-tradeab

Resolve a local time to an exact UTC instant with explicit DST ambiguity handling, business-day arithmetic, public-holiday status, and working-hour overlap across timezones. Detects ambiguous and nonexistent local times instead of silently shifting them. Declares its IANA tzdb release and per-jurisdiction holiday provenance.

Machine translation for agents between 100 languages: send text (up to 5,000 chars), target_lang and optional source_lang (auto-detect by default) and get the translation. $0.001 USDC per started 1,000 characters on Base; failures are not charged. LibreTranslate/dTelecom-style {q, source, target} also accepted at /v1/translation. M2M100 1.2B. By DevX.

Deterministic on-chain structural-safety report for a Base address: contract detection, verification status, proxy/owner/privileged-function analysis. Behavioral signals (age, activity, funding) not yet included. Keyless x402-native payment — no signup, no API key. Score 0-100, higher = structurally safer. See free real samples before paying: GET /samples/index.json

Recase a string into camelCase, PascalCase, snake_case, kebab-case, CONSTANT_CASE, dot.case, path/case, Sentence case, Capital Case, Train-Case, no case, Title Case, or plain UPPER/lower — the word boundaries are detected regardless of the input casing.

Ask anything about crypto: real-time crypto insights, token analytics, trend detection and market intelligence.

Parse an SMPTE timecode (HH:MM:SS:FF, or HH:MM:SS;FF for drop-frame) at a given frame rate into its components, total frame count and real-time seconds. Drop-frame is auto-detected from a ';' separator unless dropFrame is set. Out-of-range or non-existent drop-frame values are a clean 400.

Returns Companies using a specific Technology as a list of Technology Detections, ordered by the `first_seen_at`, descending.<br><br>E.g. using this endpoint one can get a list of all companies using HubSpot or any other of the 240K technologies PredictLeads tracks.<br><br>The specific Technology ID can be obtained by querying the **<a href="https://docs.predictleads.com/api_endpoints/technologies

Replace values at exactly named object keys recursively; no secret detection and no guarantee that unlisted fields are safe to share.

Text Moderation pii: Detect personal data (email, phone, SSN, credit card with Luhn check, IP, IBAN, crypto address, name, address, date of birth, passport) with character offsets; optional redaction to [TYPE].

Probe an MCP/agent endpoint: liveness, latency, tool count, auth metadata, price-challenge detection. Real MCP initialize handshake, not a ping.

Detect prediction market outcome mispricings — find markets where real-world outcome is already determined but market price is wrong