Animica

Random bytes from the Animica node's randomness service with a signed digest attestation and an entropy-health report, 1..1024 bytes per draw. Current trust model, stated up front and in every response: the entropy source is a software CSPRNG (os.urandom) and the signer is a software key, so source.is_quantum=false and attestation.attested=false today; those fields flip truthfully if a hardware/quantum provider is ever connected. The free catalog entry carries the same live entropy block, so you can check the source before paying.

Standard-tier chat/completions (OpenAI-compatible), served by Animica's network. Cheaper than priority, standard latency — priced at 3x settlement gas.

OpenAI-compatible chat completions with capacity-gated admission, no per-IP rate limit, direct upstream routing and a 180 s timeout. Only sold while enough workers are live-serving; disabled by default.

The ANM/USD reference price this operator actually uses, attested at a specific chain head. Quoted from the NonKYC BID (the conservative side), with the observation timestamp and feed age included so you can judge freshness yourself. NOTE: no signing key is configured on this gateway, so attestations are returned UNSIGNED (signed:false). The price is still the live one we use; it simply carries no cryptographic attestation. A stale or indicative feed makes this product unavailable rather than quoting a dead rate.

A ranked snapshot of ANM holders at a pinned chain height — up to 1000 accounts with exact nANM balances as decimal strings, plus the total address count and the head it was taken at. The shape airdrops, governance weighting and distribution analysis actually need, in one call instead of thousands of balance lookups.

A point-in-time read of the Animica mempool and network: pending transaction count and ids, total pending bytes, the age of the oldest entry, chain head, peer count and network hashrate. What a bot watching for congestion, fee conditions or inclusion timing needs, without running a node.

Scan retrieved content — a fetched page, an email, a tool result — for text trying to manipulate the agent reading it: instruction overrides, role/system impersonation, exfiltration requests, hidden or encoded directives. Returns a risk score, the suspicious spans verbatim, and the technique observed. Deterministic pattern checks run ALONGSIDE the model so a blatant "ignore all previous instructions" is caught even if the model misses it, and the two signals are reported separately rather than blended into one number you cannot interrogate. Priced per call. For high-volume use, buy prepaid credits (POST /x402/credits/buy) and send X-Animica-Credits on each request — no settlement, no gas, no per-call payment round trip. Paying in ANM on the animica:1 lane is also ~25% cheaper because we sponsor no gas there.

Assign text to one of YOUR labels — spam, sentiment, urgency, topic, ticket category, lead quality, intent, moderation — with a confidence and a one-line rationale. The label is constrained to your list by schema enum, so it cannot return a category you did not define. Priced per call. For high-volume use, buy prepaid credits (POST /x402/credits/buy) and send X-Animica-Credits on each request — no settlement, no gas, no per-call payment round trip. Paying in ANM on the animica:1 lane is also ~25% cheaper because we sponsor no gas there.

Store up to 1024 KiB and get back a content commitment you can hand to anyone. The blob is erasure-coded across shards in the Animica data-availability layer and is retrievable for free by commitment at GET /x402/blob/{commitment}, with a Merkle inclusion proof. Useful when an agent needs to hand another agent a durable, addressable artifact rather than a link that rots. Stored blobs are PUBLIC to anyone holding the commitment — do not put secrets here.

Full transaction history for one Animica account (anim1… bech32m), served from the gateway's own head-following index with a stable cursor — up to 500 rows per call, amounts in nANM as decimal strings. The free per-block and per-tx APIs stay free; single-address lookups on explorer.animica.org stay free too (they scan at most 250 blocks per call).

Balances for up to 500 Animica accounts in ONE batched call, head-pinned, amounts in nANM as exact decimal strings. Single-address balance lookups stay free on the public API — this sells the batch.

Block-range and transaction-range exports from the Animica L1 (chunked reads, cursor pagination, NDJSON/JSON, gzip). Caps per request: 1000 blocks, 10000 tx records, 16000000 bytes. The free per-block APIs stay free.

Embed up to 256 texts in ONE call with all-MiniLM-L6-v2 (384-dimensional, normalised for cosine similarity), up to 8192 characters each. Sold as a batch on purpose: a single settlement costs real gas, so a per-string price would be almost entirely payment overhead — one call, many vectors, is the only honest shape for this. This is a small fast local sentence-transformer, not a frontier embedding model; the model id rides on every response so a future change is visible rather than silent.

Audit whether a website can actually be read, quoted and cited by AI agents, and get a prioritised fix list. Probes the homepage as each real AI crawler (GPTBot, OAI-SearchBot, ChatGPT-User, ClaudeBot, Claude-User, PerplexityBot, CCBot, meta-externalagent, Amazonbot) to catch the 429s, 403s and soft-404s that make a live site invisible; parses robots.txt per agent including the robots-only training tokens Google-Extended and Applebot-Extended; and checks llms.txt, structured data, machine-readable endpoints (OpenAPI, MCP, x402, well-known), sitemap, canonical and title/meta, plus how much of the page survives with JavaScript off. Fully deterministic — no model is called, so the same site scores the same twice. It does NOT ask ChatGPT or Perplexity whether they have heard of your brand; it measures the inputs that decide whether they can.

Anchor a SHA-256 digest (or a small document, which we hash for you) into the Animica data-availability layer and get back a commitment plus a Merkle inclusion proof anyone can check without trusting us. The record is erasure-coded across shards and committed to a Merkle root; the response carries the commitment, the leaf index and sibling path, and the chain head height and hash observed at the moment of anchoring. Verification is FREE and permanent at GET /x402/notarize/verify/{commitment} — a notarisation nobody can check would be worth nothing. Note precisely what this proves: the record is committed in this node's DA tree under the returned commitment, alongside the head it observed. That is not the same as a timestamp signed by the whole consensus set, and the response says so rather than blurring the two.

Verify a post-quantum signature (ML-DSA-65 / Dilithium3 / SPHINCS+ as enabled by chain policy) against a message and public key, using the same verifier the Animica chain itself uses to admit transactions. Useful when an agent must check a PQ-signed attestation without shipping a post-quantum library of its own. A signature that does not verify is a SUCCESSFUL call answering ok:false - you paid for the answer, and "invalid" is an answer. Only an inability to perform the check is treated as a failure.

Give a URL and a question; get an answer grounded in that page, WITH the passages the answer was drawn from. We fetch the page, chunk it, embed the chunks and your question, retrieve the closest 4 passages and answer from those only. Nothing is stored — it is a one-shot pipeline, not an index. The retrieved passages come back with the answer so you can check it rather than trust it, and when nothing on the page is relevant the product says so instead of letting the model improvise. Reaches the public internet only, with the same SSRF protections as the fetch product.

Fetch a public web page and return clean readable text with its title and metadata - the form a model can actually use, with navigation, scripts and styling removed. Follows up to 3 redirects, caps the download at 2000000 bytes and gives up after 15000ms. Reaches the PUBLIC internet only: the hostname is resolved and every redirect hop re-checked against private, loopback, link-local and CGNAT ranges, so this cannot read internal services or cloud metadata. Returns raw HTML too on request.

Statistical analysis of any segment of the x402 economy, computed over a merged index of Coinbase Bazaar and 402index overlaid with our own unpaid 402 probes. Returns the full price distribution (min, p10, p25, median, p75, p90, max, mean), demand discounted by payer concentration, what share of the segment can actually be invoked, the network and asset split, host concentration, liveness, and — once history exists — the trend. Every figure is computed in code from named rows; the interpretation is written by Animica's own AICF inference network and every sentence containing a number not in the computed facts is deleted before delivery. Directories publish rows; this publishes the distribution.

Competitive position for a single x402 resource. Give a URL that is listed anywhere in the merged index and get back what the index holds on it (including any correction our own unpaid 402 probe made to its advertised price), the peer set derived from its own description, and where it ranks among those peers on price, call volume, payer diversity and callability — plus the mirrors of itself that are listed as separate services. Useful whether the endpoint is yours or one you are about to buy from. Interpretation is written by Animica's own AICF inference network; every figure is computed in code.

Price positioning for an x402 service. Describe what you sell (and optionally what you charge) and get the comparable set — named, counted, and listed — plus your exact percentile within it, the full decile ladder of what those comparables charge, how much demand sits in each price band, and a suggested band derived from the comparables rather than asserted. Refuses to compute a percentile when too few real comparables exist, because a confident percentile over four loosely-related rows is worse than no answer. Interpretation is written by Animica's own AICF inference network; every figure is computed in code.

Live ANM monetary state from the Animica L1 itself: supply and account count, the block reward the node will actually credit (not the published split, which is wrong by 2x), measured blocks per day, daily emission in ANM and as a percent of supply, network hashrate, and — when the ANM book is fresh — total network-wide mining revenue in USD per day. Buy before holding, quoting or mining ANM. Deterministic, no model. Charges nothing when the chain cannot be read.

Whether the Animica L1 is healthy enough to transact on right now: head height and hash, sync lag (blocks vs headers), difficulty and PoIES theta, measured network hashrate with its window and estimator named, mempool depth, and peer count as a RANGE across every source — because four sources on this node disagree at the same instant. Publishes NO peer IP, peer id or relay attribution. Read live, no model, free when the chain cannot be read.

Which Animica signature scheme you can actually build on. The chain reports three schemes enabled, but two are deprecated commitment stubs the node refuses to make keys for — and 4098 wallets are already stranded on one. Returns the live scheme list plus both id spaces (consensus schemeId and PQ algorithm id 0x1003), key and signature byte sizes, and a usable/not verdict per scheme with its evidence. Buy before generating a key or writing a signer. No model. Free when the chain cannot be read.

Live capacity of the Animica compute network in one call: which GPU render kinds have miners online right now and how many, whether a registered AICF worker is serving inference, and the chain height it was measured at. Buy it BEFORE you pay for a render or an inference call, so you pick a servable kind instead of learning from a job nothing claims. Anything we could not reach reports available:false with a reason: counts are never guessed and never a zero we did not measure.

Buy bulk crawl access to one domain that publishes Paid Crawl terms. $0.010 buys $0.010 worth of that site's pages — the site sets the per-page rate, so the request budget is the pass value divided by that rate. ONE settlement covers the whole budget: per-page settlement is impossible at these prices because Base gas would exceed the fee. Send the returned token back as "X-Crawl-Pass: anmcp_...". Site operators pay nothing for Paid Crawl; crawlers do.

Pull people, organizations, locations, dates, amounts, products, URLs, emails, phone numbers and wallet addresses out of text, each with the exact surface form as it appeared so you can locate it. Types are constrained by schema, so the result is always one of the known kinds. Priced per call. For high-volume use, buy prepaid credits (POST /x402/credits/buy) and send X-Animica-Credits on each request — no settlement, no gas, no per-call payment round trip. Paying in ANM on the animica:1 lane is also ~25% cheaper because we sponsor no gas there.

Give a task and a budget; get the result, a measured verification signal, and an ML-DSA-65 signed receipt over the exact bytes delivered. Routes the task to a capability and says which one ran: extract (fetch a URL and read it), research, predict (with the live prediction-market price where one matches), chain (Animica L1 facts), code, summarize, or a direct answer. HONEST ABOUT VERIFICATION: there is one inference backend today, so multi-sample checking measures SELF-CONSISTENCY, not independent providers — the response states provider_count, whether providers were distinct, and what the agreement number means. Confidence is measured from real agreement or reported as null; it is never a decorative number. The receipt is signed with the same post-quantum scheme this chain admits transactions with, the public key ships with it, and the result can optionally be anchored on-chain with a free permanent proof.

Turn messy text — a web page, email, invoice, receipt, job posting, contract clause, product description — into strict JSON matching a schema YOU supply. The schema is enforced in code, not merely requested: output is parsed, validated, repaired once against the exact validation error, and the call FAILS with the violated constraint rather than returning unvalidated JSON. Fields not present in the input come back null instead of a plausible invention. Priced per call. For high-volume use, buy prepaid credits (POST /x402/credits/buy) and send X-Animica-Credits on each request — no settlement, no gas, no per-call payment round trip. Paying in ANM on the animica:1 lane is also ~25% cheaper because we sponsor no gas there.

Ask a yes/no question about the future and get three things: an independent model probability, the live price of the matching prediction market when one exists, and — the point of the product — the whole record anchored into the Animica data-availability layer with a commitment anyone can verify FREE and permanently at GET /x402/forecast/{commitment}. Anyone can claim afterwards that they called it; this proves what was believed beforehand. Both numbers are always returned side by side: the model is NOT claimed to beat the market, and our own accuracy against every resolved market is published free at GET /x402/forecast/calibration, including when we lose. If the record cannot be anchored the call fails rather than selling an unanchored prediction.

Emit the actual files that fix a site's AI legibility, ready to deploy: a real /llms.txt, a /robots.txt that stops excluding AI crawlers (with a unified diff against the current one), and a JSON-LD block for the page <head>. Nothing is invented: every link in the llms.txt is discovered on the site and then FETCHED to confirm it answers 200, with failures dropped and counted, and every JSON-LD field is copied from something the page actually declares — anything that cannot be grounded is omitted rather than guessed. A small model writes only the prose, from the titles and descriptions we extracted; any URL it returns that is not in the verified set is discarded. Pairs with /x402/geo/audit: audit tells you what is wrong, this hands you the files. Re-run the audit afterwards to confirm the score moved.

Hand over malformed model output and a JSON Schema; get back JSON that provably satisfies it. Tries a pure-parse fast path FIRST and returns immediately when the input is already valid — you are not charged a model call for text that merely looked broken. Otherwise a small model repairs it and the result is validated in code before it is returned. Priced per call. For high-volume use, buy prepaid credits (POST /x402/credits/buy) and send X-Animica-Credits on each request — no settlement, no gas, no per-call payment round trip. Paying in ANM on the animica:1 lane is also ~25% cheaper because we sponsor no gas there.

Search the whole x402 economy and get a ranked shortlist of services that can do a job, within a budget. Merges Coinbase Bazaar and 402index into one index keyed by canonical resource URL, then answers the three questions a directory does not: CAN this be invoked (only about 5% publish a call spec, and that is a first-class field here, not something you infer from missing keys), is the price REAL (one live listing advertises ten billion dollars a call — implausible prices are excluded as bad data and reported, never silently rewritten), and is the demand real (total calls are trivially inflated by a merchant calling itself, so volume is credited only in proportion to distinct paying counterparties). Every result explains its own score, and the weights are returned so you can disagree with them. Lexical matching, no model: the same query against the same index ranks the same way.

Call any x402 endpoint WITHOUT paying and get back what its own 402 challenge says: the real price, asset, network, payTo address, scheme, timeout, and the request schema when the merchant publishes one. Directory listings are written once and go stale; this is the merchant's own current statement of the terms you are about to accept. It also distinguishes the cases a directory cannot: a resource that answers 402 properly, one that is listed as paid but answers WITHOUT payment, one that is listed but gone, and one that simply fails. No payment header is ever sent, so this cannot settle anything, and unknown endpoints are probed with GET — never a write verb with a guessed body.

Which Animica workloads a machine can earn on, in one poll. Per-kind GPU render supply, with the kinds NOBODY serves flagged as supply gaps; the proof-of-work lane off an Animica node (difficulty, thetaMicro, hashrate with unit+window, mempool depth); VRAM floor and deps per kind from the reference miner client; probed enrolment endpoints. Buy it to decide where to point GPU or hash capacity, or to size a box. Empty queues read empty, unreadable ones unreadable. No observable lane, no charge.

Uniform integers in [min, max] derived from ONE verified randomness draw by rejection sampling (never modulo bias) — up to 1000 per request. The response carries the raw draw, the node's source/health/attestation and the exact rule, so you can recompute every integer offline.

Give a query and up to 50 candidate snippets; get the top N by relevance with a score and a one-line reason. Indexes are returned so you can map straight back to your own objects without string matching. Priced per call. For high-volume use, buy prepaid credits (POST /x402/credits/buy) and send X-Animica-Credits on each request — no settlement, no gas, no per-call payment round trip. Paying in ANM on the animica:1 lane is also ~25% cheaper because we sponsor no gas there.

Give a goal and the actions available to you; get back the single best action, its arguments, a confidence and a reason. The chosen action is constrained by schema enum to your list, so it can never name a tool you do not have. Built for the decision an agent makes before spending a frontier-model call — including the honest answer that it should escalate. Priced per call. For high-volume use, buy prepaid credits (POST /x402/credits/buy) and send X-Animica-Credits on each request — no settlement, no gas, no per-call payment round trip. Paying in ANM on the animica:1 lane is also ~25% cheaper because we sponsor no gas there.

Describe a goal and a budget; get back a concrete plan of real x402 calls that would accomplish it — exact URLs, HTTP methods, request shapes, per-step prices from the merchants' own 402 challenges where we have verified them, alternates for every step, and a total you can check before committing. Only services that publish a callable request shape are planned; a step with no invokable candidate is returned as an explicit gap rather than a URL and a shrug. IT DOES NOT SPEND: this compiles the plan and stops, so you execute it with your own wallet on your own terms. It also does not claim the pipeline type-checks — roughly 5% of the x402 economy publishes schemas, so most step-to-step handoffs cannot be verified, and every edge says which it is.

Verify you are on real Animica mainnet, and which chain id to sign with. Animica answers that question with TWO numbers: the native chain is id 1, while the EVM compatibility facade reports 149 (eth_chainId 0x95, net_version "149"). A wallet porting Ethereum tooling signs with the wrong one. Returns both, the disagreement named, plus genesis hash, fork id, policy hash and the built-in consensus checkpoints a light client pins against. All read live. No model.

Give it up to 10 URLs and get back the readable text of each — titles, clean body text, navigation and scripts removed — in a single call and a single payment. The half of every retrieval pipeline an agent already has links for: search results, a sitemap, an RSS feed, a user's reading list. Pages are fetched concurrently under a deadline. Every URL you send is accounted for: the ones that could not be read come back in failed[] WITH the reason (403, timeout, paywall, no extractable text, private address), because a bad input and a broken fetcher need opposite responses from you. If not one URL could be read, you are NOT charged.

Text-to-image, rendered by an online GPU miner. Rendered by the Animica GPU-miner queue. Sold only while a miner for the requested kind is online — the catalog reports available:false and the endpoint answers 503 otherwise. Returns a job handle immediately (rendering is asynchronous); polling the returned poll_url is free. Kinds: image.

5..10 INDEPENDENT randomness draws — each one its own node call with its own signed digest attestation and health report — delivered under ONE payment of $0.004. Fewer than 5 draws is refused with a 400 that points at GET /x402/qrng/draw ($0.001 per draw), because only at 5+ draws is this genuinely cheaper per draw. Priced per draw, not per byte: if you just need many bytes under one attestation, a single 1024-byte qrng draw is cheaper — the response states which is cheaper for the call you made.

Commit now, reveal to everyone later: the paid call returns a commitment sha3_256(secret||salt) and a commit id; GET /x402/random/reveal/{commit_id} is FREE, public and idempotent, and discloses the secret, the salt, the raw draw and the node's attestation so anyone — not just the buyer — can check the commitment. Reveal delay 0..604800s. For games, sealed-bid rounds and auditable draws. Trust model, stated plainly: the commitment binds the outcome from the moment you publish it, and the reveal proves the secret came from a node-signed draw — it does NOT prove the operator discarded no draws before committing, so publish the commitment to your players before the round.

Pick k of your items with or without replacement, optionally weighted by integer weights (raffles, lotteries, sortition, A/B splits). One verified randomness draw, cumulative-weight search over a rejection-sampled uniform value, up to 10000 items. The exact rule and the raw draw ride along so you can recompute the winners.

Fisher-Yates permutation of your list (or of 1..N), up to 10000 items, derived from ONE verified randomness draw. Returns the permutation of 0-based indices plus the shuffled items, with the exact swap-and-byte-consumption rule so you can recompute it offline.

Block-production concentration on the Animica L1 over the last N blocks: every producer with its block count and share, HHI, Gini, effective producers, top-producer share and Nakamoto coefficient at 33/50/66%, plus the same over the contiguous prior window and the delta. Deterministic, no model; every figure pinned to the block range and head it was computed at. Buy before trusting this chain's finality, or to watch concentration move. Charges nothing when producer data cannot be read.

Price and feasibility for ONE specific Animica compute job before you pay for it. Send {kind, prompt, quantity} and get the exact cost in atomic units, whether a GPU miner or AICF worker is online for that exact kind right now (with probe age and chain head), the endpoint and body to POST, and a pre-check of your payload. Refuses with 503 and charges nothing when nothing can serve the work. Buy this when a failed render would cost you more than the quote.

Buy a prepaid credit voucher in ONE on-chain settlement, then spend it across every paid product on this gateway with no further settlement and no gas. Send the returned token as "X-Animica-Credits: anmc_..." (or "Authorization: Bearer anmc_...") on any paid route and the call is served from the balance instead of a payment. Credit is granted at face value plus 10% — that bonus is the Base settlement gas we do not spend on your calls 2..N, returned to you. The voucher is a BEARER secret shown exactly once and stored only as its SHA-256 hash: anyone holding it can spend it, and we cannot recover it for you. Balance and full spend history are readable for free at GET /x402/credits/balance with the token. Unused credit expires 365 days after purchase. Send the token on THIS route to top up an existing voucher instead of minting a new one.

Full-text search over what machine-payable services actually publish — their descriptions, input schemas, parameter names, prices and settlement networks — across ~18,000 x402 endpoints. Built from their own payment challenges, which a general crawler discards as a dead page: ask for a parameter name, a schema field or a network and you match services whose one-line blurb never mentions it. Filter by price ceiling, by whether the service publishes a call spec, and by payment protocol. Every response states how many documents were searched and how stale the oldest is, because an index that will not tell you its size is asking to be trusted about the one thing you cannot check. Ranked by BM25 — no model, so the same query ranks the same way twice. If the index is empty or nothing matches, you are NOT charged.

Competitors for one x402 resource or capability, over our ~18,000-service harvest of the x402 directories plus our own unpaid 402 probes. Returns the price band (min/median/p90/n) over real listings only, price provenance per rival, whether it publishes a spec an agent can call unattended, observed 30-day demand and settlement network. Never probed reads observed:false, never a guess. Buy before pricing or listing a service, or choosing an endpoint. Charges nothing when nothing matches.

Before you pay an x402 endpoint you have never called, buy what we observed when WE called it: the outcome of our own unpaid probe, when we ran it, whether it answered a correct 402 with readable payment terms, and whether it publishes a call spec you can build a request from. Never probed returns observed:false — an honest unknown, never an invented uptime figure. Optional ML-DSA-65 attestation. If we hold no record of the URL at all, you are not charged.

Deterministic mining P&L for Animica L1 at a named chain height. The per-block reward comes from the node's own live block template, not the published subsidy split (which is wrong by 2x on mainnet); the block interval and difficulty are measured, the ANM price is the NonKYC bid. Buy before pointing hardware here: it states the FIXED daily pot every miner shares, your share AFTER you join, and how much you dilute the incumbents. No model, no growth curve. Backend down = 503, no charge.

Buy bulk crawl access to one domain that publishes Paid Crawl terms. $0.100 buys $0.100 worth of that site's pages — the site sets the per-page rate, so the request budget is the pass value divided by that rate. ONE settlement covers the whole budget: per-page settlement is impossible at these prices because Base gas would exceed the fee. Send the returned token back as "X-Crawl-Pass: anmcp_...". Site operators pay nothing for Paid Crawl; crawlers do.

Buy bulk crawl access to one domain that publishes Paid Crawl terms. $1.000 buys $1.000 worth of that site's pages — the site sets the per-page rate, so the request budget is the pass value divided by that rate. ONE settlement covers the whole budget: per-page settlement is impossible at these prices because Base gas would exceed the fee. Send the returned token back as "X-Crawl-Pass: anmcp_...". Site operators pay nothing for Paid Crawl; crawlers do.

Seal any JSON in a post-quantum signed envelope: canonicalise (keys sorted, stable bytes), SHA3-256, then sign that digest with ML-DSA-65 (FIPS 204, scheme 4099) — the scheme the Animica L1 admits transactions with. Buy it when a response you emit must cross a broker, cache or reseller and still be provably unaltered. Proves INTEGRITY and ANIMICA ORIGIN, not that the content is true and not anyone's real-world identity. Key and verification are free. No signature, no charge.

Get a post-quantum ML-DSA-65 signature (FIPS 204, scheme 4099 — the same scheme the Animica L1 admits transactions with) over a document you never have to send us. Post a sha256 digest, or post a JSON payload and we canonicalise and hash it here; either way the signature covers the HASH, so the resulting envelope can be published, escrowed or put on-chain without disclosing the content. Buy this when a claim has to stay checkable by a third party years from now — a dataset manifest, a model card, an audit-log segment, a delivery receipt between two agents, the exact bytes of an agreement — because an HMAC proves nothing to anyone but its issuer and an ECDSA signature made today is retrospectively forgeable once a cryptographically-relevant quantum computer exists. You need no PQ library and no key of your own. Verification is FREE (POST /x402/pq/sign/verify) and the signing public key is FREE (GET /x402/pq/sign/pubkey), so anyone can check an envelope offline without paying us. The signed statement records whether YOU or WE computed the digest, because those are different claims. If the signer or its key is not reachable you get a 503 with the reason and are NOT charged — this endpoint has no branch that returns an unsigned envelope.

Find out why an Animica transaction would be rejected BEFORE you broadcast it. Decodes your canonical CBOR {body, sigs} envelope, returns the normalised fields and the CANONICAL HASH you are about to sign, then asks the node whether it would admit the transaction and why not. Both calls are read-only: nothing is inserted into the mempool and nothing is broadcast. Built for people writing Animica signers, where the alternative feedback loop is broadcast-and-see. No model.