openapi: 3.2.0 info: contact: email: support@getanyapi.com description: 'Any API, one wallet, USD, no subscriptions. Each API is a single discovered operation: send the normalized input, get a normalized result. Prices and payment outcomes are reported in USD; each operation documents the settlement policy for its enabled payment rails.' title: Any Google API version: 1.0.0 x-guidance: 'Use each operation''s published method and path with its normalized JSON input (see the operation requestBody). Before setting a client or tool timeout, GET /v1/apis/{sku} and inspect its trailing-30-day latency p50/p95/p99 and sample; p99 is an observation, not a maximum. To pay: send your AnyAPI key (Authorization: Bearer, billed from your USD wallet). For operations that advertise an inline rail, pay per call inline with x402 - call with no key, receive HTTP 402 with a PAYMENT-REQUIRED header, then retry with the PAYMENT-SIGNATURE header (base, no account needed); or pay per call inline with MPP (Machine Payments Protocol) - call with no key, receive HTTP 402 with a WWW-Authenticate: Payment challenge, then retry with the Authorization: Payment header (tempo, no account needed). x402 settles after execution; execution failure is reported as released; mpp settles before execution; execution failure is reported as charged_undelivered.' servers: - url: https://api.getanyapi.com security: - bearerAuth: [] - apiKeyAuth: [] tags: - name: Google x-group: Google paths: /v1/run/google.ai_mode: post: description: 'Ask Google AI Mode a prompt and receive the cited answer it generates. AI Mode composes the answer at search time, so repeat calls on one prompt can differ in wording and in which sources are cited. **Price:** \$1.26 per 1,000 requests (flat per request - same cost regardless of results returned). **Routing:** 2 independent lanes are available in execution order. A failed attempt uses the next lane only when the request''s payment authorizes that lane''s price; the normalized schema stays the same. **Catalog:** Google AI Mode pricing and uptime - live USD price, lane routing, and measured 30-day uptime. Every Google endpoint.' operationId: google_ai_mode parameters: - description: Optional. Comma-separated keys (dotted paths like `author.name` descend into nested objects) to keep on each result item. Keys are matched relative to each result item after the `data`/`items` envelope is unwrapped, not against the top-level response envelope, so use `jq` to reshape the whole envelope. Shrinks the response without changing cost. in: query name: fields required: false schema: type: string - description: Optional. Cap the number of result rows returned; a `_truncated` note reports how many were withheld so you can page via the API's own `limit`. Does not change cost. in: query name: max_items required: false schema: minimum: 0 type: integer - description: Optional. Return only a structural outline (top-level keys, item counts, and per-field byte sizes) instead of the full data. Does not change cost. in: query name: summary required: false schema: type: boolean - description: 'Optional. A jq expression applied to the result envelope; its output replaces `output` (multiple outputs collect into an array). Reshape freely, e.g. `jq=.data | {title, description, md: .markdown[:3500]}`. Runs sandboxed with a 250ms / 2MB budget; on failure the full result is returned with a `jqError`. Does not change cost.' in: query name: jq required: false schema: type: string - description: Optional. The most you are willing to pay for this one request, in US dollars (for example 0.05). Any route that would charge more than this is not used, so a request only ever runs on something you can afford. If nothing is available at or below your amount, the request is refused before it runs, nothing is charged, and the message tells you the cheapest price per request so you can raise it. Leave it out to accept the normal price. in: query name: max_cost_usd required: false schema: type: string - description: Optional wallet idempotency key, scoped to this customer for 24 hours. When the gateway honors the key, this synchronous in-process execution can continue after the caller disconnects, bounded by its execution deadline. A completed replayable result charges normally exactly once and can be replayed without another provider run or charge. A pending duplicate returns 409 idempotency_in_progress; reuse with different request semantics returns 409 idempotency_conflict. in: header name: Idempotency-Key required: false schema: maxLength: 255 minLength: 1 type: string requestBody: content: application/json: example: prompt: What is AnyAPI at getanyapi.com, and what does it offer? schema: additionalProperties: false example: prompt: What is AnyAPI at getanyapi.com, and what does it offer? properties: preferLatencyUnderMs: description: 'Optional; omit it and routing is unchanged, with the cheapest source serving. Prefer sources whose typical response time (median over the trailing 30 days, as published on this endpoint''s lane health) is under this many milliseconds; among those, the cheapest serves. This can raise your price: when the cheapest source misses the target, a faster and dearer one serves, and you are quoted and charged its price. If no source is that fast the request is still served, by whichever source offers the best speed for its price - it is never refused for being slow. Sources we have not timed are tried last. This is a preference, not a guarantee: the median describes past requests and is not a ceiling on this one, and it excludes any wait this request itself asks for. On a paginated walk it applies to the first page only: later pages stay with the source that page chose, at the price it was quoted.' minimum: 1 type: integer prompt: description: The question or prompt to answer with Google AI Mode. minLength: 1 type: string required: - prompt title: Google AI Mode input type: object required: true responses: '200': content: application/json: schema: properties: costUsd: description: USD charged on the original run. On a replay this value is echoed for parity; the replay itself is free. type: number hint: description: 'Optional one-line nudge, absent when there is nothing to say. `large_result:` suggests the fields/max_items/summary/jq controls for a big response. `paging_unavailable:` means this result came from a source that cannot return a `nextCursor`, so it may be INCOMPLETE and cannot be continued - re-run with `requireCursor: true` to be served only by a source that can page, which may cost more per request.' type: string items: description: Number of result rows returned. For per-result SKUs the per-item cost is charged against this count; for input-priced SKUs the charge is per submitted input, independent of this count. type: integer jqError: description: Present only when a `jq` expression failed; `output` then carries the full unshaped result and this explains why the reshape did not apply. type: string output: anyOf: - additionalProperties: false properties: data: description: The normalized answer, or null when no answer was found. Google composes the answer at search time rather than serving a stored one, so the same prompt can return different wording and a different set of citations on each call. Treat a response as one observation of the answer, not as a stable document. oneOf: - type: 'null' - additionalProperties: false properties: answer: description: The answer as plain text, as Google generated it for this search. Length and coverage vary between searches on the same prompt. Populated whenever the provider has data for the entity. type: string x-anyapi-must-populate: true answerMarkdown: description: The answer in Markdown when Google returns a Markdown rendering, otherwise the same text as answer. Populated whenever the provider has data for the entity. type: string x-anyapi-must-populate: true citations: description: The sources Google cited for this search, in the order it returned them. Google recomputes the set per search, so counts and membership vary between calls on the same prompt. items: properties: title: description: The cited source title. type: string url: description: The cited source URL. type: string required: - title - url type: object type: array prompt: description: The prompt answered by the upstream search experience. Populated whenever the provider has data for the entity. type: string x-anyapi-must-populate: true required: - prompt - answer - answerMarkdown - citations type: object found: description: Whether Google AI Mode returned an answer. type: boolean required: - found - data title: Google AI Mode output type: object - type: 'null' description: Normalized output, or null when the replay payload was not retained. provider: description: Always "AnyAPI". type: string replayed: description: True when this response replays the durable result of an earlier run without billing or upstream execution. type: boolean resultId: description: Opaque handle to the full unshaped result, cached ~15 min. Re-shape it for free (fields/max_items/summary/jq) via GET /v1/results/{id}, no re-billing. Absent when the result was too large to cache. type: string required: - output - provider - costUsd - items - replayed type: object description: Normalized result. headers: Idempotency-Replayed: description: Present only when the request carried an Idempotency-Key this gateway honored. "false" means the key was claimed and the run executed freshly; "true" means it replayed a prior identical run and was NOT charged again. Absent means either no key was sent or this gateway does not honor keys, so a retry could be charged twice. schema: enum: - 'false' - 'true' type: string PAYMENT-RESPONSE: description: Inline x402 payment receipt. schema: type: string Payment-Receipt: description: Inline mpp payment receipt. schema: type: string X-PAYMENT-RESPONSE: description: Inline x402 payment receipt. schema: type: string '400': content: application/json: schema: $ref: '#/components/schemas/Error' description: Input did not match the schema, or the upstream rejected the request. '401': content: application/json: schema: $ref: '#/components/schemas/Error' description: Missing or invalid API key. '402': content: application/json: schema: oneOf: - $ref: '#/components/schemas/PaymentRequiredError' - $ref: '#/components/schemas/TrialCapReached' application/problem+json: schema: properties: challengeId: minLength: 1 type: string status: const: 402 type: integer required: - status - challengeId type: object description: Insufficient wallet balance or spend cap, or an inline-payment challenge. Agent trials receive a trial_cap_reached continuation with a live RFC 8628 device authorization. headers: PAYMENT-REQUIRED: description: Inline x402 payment challenge. schema: type: string WWW-Authenticate: description: Inline mpp payment challenge. schema: type: string '404': content: application/json: schema: $ref: '#/components/schemas/Error' description: Unknown SKU, or the upstream had no matching resource. '409': content: application/json: schema: $ref: '#/components/schemas/Error' description: 'Idempotency collision. Codes: idempotency_in_progress means retry after the Retry-After delay; idempotency_conflict means use a new key for different request semantics; idempotency_needs_review means stop retrying and contact AnyAPI support with X-Anyapi-Request-Id.' '422': content: application/json: schema: $ref: '#/components/schemas/Error' description: The request was well formed but this API cannot serve the requested target site; the body's `alternatives` array lists the AnyAPI APIs that can. '429': content: application/json: schema: $ref: '#/components/schemas/Error' description: The upstream was rate limited; inspect `payment` when present before deciding whether to retry. '500': content: application/json: schema: $ref: '#/components/schemas/Error' description: An internal error prevented completion. Inspect payment when present before retrying. '502': content: application/json: schema: $ref: '#/components/schemas/Error' description: Every provider failed with a server/transport error, or inline payment settlement could not be confirmed; inspect `payment` when present before retrying. security: - bearerAuth: [] - apiKeyAuth: [] - {} summary: Google AI Mode tags: - Google x-mint: href: /api-reference/google/ai-mode x-payment-info: price: currency: USD max: '0.003360' min: '0.001260' mode: dynamic protocols: - x402: asset: '0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913' network: eip155:8453 payTo: '0x754BD2A6799e766B1E2b9f1Ae7fBdC0CeD96ef13' scheme: exact - mpp: asset: '0x20c000000000000000000000b9537d11c60e8b50' currency: USD intent: charge method: tempo network: tempo payTo: '0x754BD2A6799e766B1E2b9f1Ae7fBdC0CeD96ef13' rail: tempo scheme: exact rails: - headers: challenge: - PAYMENT-REQUIRED credential: - PAYMENT-SIGNATURE - X-PAYMENT receipt: - PAYMENT-RESPONSE - X-PAYMENT-RESPONSE pricing: currency: USD max: '0.003360' min: '0.001260' minimumUsd: '0.001000' mode: dynamic quoteBasis: floor protocol: asset: '0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913' network: eip155:8453 payTo: '0x754BD2A6799e766B1E2b9f1Ae7fBdC0CeD96ef13' scheme: exact rail: x402 settlement: onExecutionFailure: released onUnknownFinalization: indeterminate timing: after_execution - headers: challenge: - WWW-Authenticate credential: - Authorization receipt: - Payment-Receipt pricing: currency: USD max: '0.003360' min: '0.001260' minimumUsd: null mode: dynamic quoteBasis: funded_ceiling protocol: asset: '0x20c000000000000000000000b9537d11c60e8b50' currency: USD intent: charge method: tempo network: tempo payTo: '0x754BD2A6799e766B1E2b9f1Ae7fBdC0CeD96ef13' rail: tempo scheme: exact rail: mpp settlement: onExecutionFailure: charged_undelivered onVerificationError: indeterminate timing: before_execution /v1/run/google.ai_overview: post: description: 'Ask Google Search a prompt and receive the AI Overview it generated for that scrape, with every source Google listed. Google regenerates the overview per search, so the same prompt can return different wording and a different source list. **Price:** \$1.80 per 1,000 requests (flat per request - same cost regardless of results returned). **Routing:** 2 independent lanes are available in execution order. A failed attempt uses the next lane only when the request''s payment authorizes that lane''s price; the normalized schema stays the same. **Catalog:** Google AI Overview pricing and uptime - live USD price, lane routing, and measured 30-day uptime. Every Google endpoint.' operationId: google_ai_overview parameters: - description: Optional. Comma-separated keys (dotted paths like `author.name` descend into nested objects) to keep on each result item. Keys are matched relative to each result item after the `data`/`items` envelope is unwrapped, not against the top-level response envelope, so use `jq` to reshape the whole envelope. Shrinks the response without changing cost. in: query name: fields required: false schema: type: string - description: Optional. Cap the number of result rows returned; a `_truncated` note reports how many were withheld so you can page via the API's own `limit`. Does not change cost. in: query name: max_items required: false schema: minimum: 0 type: integer - description: Optional. Return only a structural outline (top-level keys, item counts, and per-field byte sizes) instead of the full data. Does not change cost. in: query name: summary required: false schema: type: boolean - description: 'Optional. A jq expression applied to the result envelope; its output replaces `output` (multiple outputs collect into an array). Reshape freely, e.g. `jq=.data | {title, description, md: .markdown[:3500]}`. Runs sandboxed with a 250ms / 2MB budget; on failure the full result is returned with a `jqError`. Does not change cost.' in: query name: jq required: false schema: type: string - description: Optional. The most you are willing to pay for this one request, in US dollars (for example 0.05). Any route that would charge more than this is not used, so a request only ever runs on something you can afford. If nothing is available at or below your amount, the request is refused before it runs, nothing is charged, and the message tells you the cheapest price per request so you can raise it. Leave it out to accept the normal price. in: query name: max_cost_usd required: false schema: type: string - description: Optional wallet idempotency key, scoped to this customer for 24 hours. When the gateway honors the key, this synchronous in-process execution can continue after the caller disconnects, bounded by its execution deadline. A completed replayable result charges normally exactly once and can be replayed without another provider run or charge. A pending duplicate returns 409 idempotency_in_progress; reuse with different request semantics returns 409 idempotency_conflict. in: header name: Idempotency-Key required: false schema: maxLength: 255 minLength: 1 type: string requestBody: content: application/json: example: prompt: How does photosynthesis work? schema: additionalProperties: false example: prompt: How does photosynthesis work? properties: preferLatencyUnderMs: description: 'Optional; omit it and routing is unchanged, with the cheapest source serving. Prefer sources whose typical response time (median over the trailing 30 days, as published on this endpoint''s lane health) is under this many milliseconds; among those, the cheapest serves. This can raise your price: when the cheapest source misses the target, a faster and dearer one serves, and you are quoted and charged its price. If no source is that fast the request is still served, by whichever source offers the best speed for its price - it is never refused for being slow. Sources we have not timed are tried last. This is a preference, not a guarantee: the median describes past requests and is not a ceiling on this one, and it excludes any wait this request itself asks for. On a paginated walk it applies to the first page only: later pages stay with the source that page chose, at the price it was quoted.' minimum: 1 type: integer prompt: description: The question or prompt to answer with a Google AI Overview. minLength: 1 type: string required: - prompt title: Google AI Overview input type: object required: true responses: '200': content: application/json: schema: properties: costUsd: description: USD charged on the original run. On a replay this value is echoed for parity; the replay itself is free. type: number hint: description: 'Optional one-line nudge, absent when there is nothing to say. `large_result:` suggests the fields/max_items/summary/jq controls for a big response. `paging_unavailable:` means this result came from a source that cannot return a `nextCursor`, so it may be INCOMPLETE and cannot be continued - re-run with `requireCursor: true` to be served only by a source that can page, which may cost more per request.' type: string items: description: Number of result rows returned. For per-result SKUs the per-item cost is charged against this count; for input-priced SKUs the charge is per submitted input, independent of this count. type: integer jqError: description: Present only when a `jq` expression failed; `output` then carries the full unshaped result and this explains why the reshape did not apply. type: string output: anyOf: - additionalProperties: false properties: data: description: The AI Overview Google generated for this scrape, or null when Google returned no overview. Google regenerates the overview on each search, so two calls with the same prompt can return different wording and a different set of sources. Treat a response as a timestamped observation of what Google served, not as a stable record. oneOf: - type: 'null' - additionalProperties: false properties: answer: description: The AI Overview answer as plain text, as generated for this scrape. Populated whenever the provider has data for the entity. type: string x-anyapi-must-populate: true answerMarkdown: description: The same answer in Markdown, preserving the headings and lists Google rendered. Populated whenever the provider has data for the entity. type: string x-anyapi-must-populate: true citations: description: Every source Google listed for this overview, in Google's own order. This is the full list behind the overview, not only the few sources Google renders inline before the list is expanded, so it is routinely longer than what a reader sees at a glance. items: properties: index: description: Google's position for this source within the overview, starting at 1. type: integer title: description: The cited source title as Google presented it, which may carry a trailing date and snippet. type: string url: description: The cited source URL. type: string required: - title - url type: object type: array prompt: description: The prompt Google answered. Populated whenever the provider has data for the entity. type: string x-anyapi-must-populate: true scrapedAt: description: When this overview was captured, ISO 8601 UTC. Because Google regenerates the overview per search, this identifies which generation the other fields describe. type: string required: - prompt - answer - answerMarkdown - citations type: object found: description: Whether Google Search returned an AI Overview for this prompt. type: boolean required: - found - data title: Google AI Overview output type: object - type: 'null' description: Normalized output, or null when the replay payload was not retained. provider: description: Always "AnyAPI". type: string replayed: description: True when this response replays the durable result of an earlier run without billing or upstream execution. type: boolean resultId: description: Opaque handle to the full unshaped result, cached ~15 min. Re-shape it for free (fields/max_items/summary/jq) via GET /v1/results/{id}, no re-billing. Absent when the result was too large to cache. type: string required: - output - provider - costUsd - items - replayed type: object description: Normalized result. headers: Idempotency-Replayed: description: Present only when the request carried an Idempotency-Key this gateway honored. "false" means the key was claimed and the run executed freshly; "true" means it replayed a prior identical run and was NOT charged again. Absent means either no key was sent or this gateway does not honor keys, so a retry could be charged twice. schema: enum: - 'false' - 'true' type: string PAYMENT-RESPONSE: description: Inline x402 payment receipt. schema: type: string Payment-Receipt: description: Inline mpp payment receipt. schema: type: string X-PAYMENT-RESPONSE: description: Inline x402 payment receipt. schema: type: string '400': content: application/json: schema: $ref: '#/components/schemas/Error' description: Input did not match the schema, or the upstream rejected the request. '401': content: application/json: schema: $ref: '#/components/schemas/Error' description: Missing or invalid API key. '402': content: application/json: schema: oneOf: - $ref: '#/components/schemas/PaymentRequiredError' - $ref: '#/components/schemas/TrialCapReached' application/problem+json: schema: properties: challengeId: minLength: 1 type: string status: const: 402 type: integer required: - status - challengeId type: object description: Insufficient wallet balance or spend cap, or an inline-payment challenge. Agent trials receive a trial_cap_reached continuation with a live RFC 8628 device authorization. headers: PAYMENT-REQUIRED: description: Inline x402 payment challenge. schema: type: string WWW-Authenticate: description: Inline mpp payment challenge. schema: type: string '404': content: application/json: schema: $ref: '#/components/schemas/Error' description: Unknown SKU, or the upstream had no matching resource. '409': content: application/json: schema: $ref: '#/components/schemas/Error' description: 'Idempotency collision. Codes: idempotency_in_progress means retry after the Retry-After delay; idempotency_conflict means use a new key for different request semantics; idempotency_needs_review means stop retrying and contact AnyAPI support with X-Anyapi-Request-Id.' '422': content: application/json: schema: $ref: '#/components/schemas/Error' description: The request was well formed but this API cannot serve the requested target site; the body's `alternatives` array lists the AnyAPI APIs that can. '429': content: application/json: schema: $ref: '#/components/schemas/Error' description: The upstream was rate limited; inspect `payment` when present before deciding whether to retry. '500': content: application/json: schema: $ref: '#/components/schemas/Error' description: An internal error prevented completion. Inspect payment when present before retrying. '502': content: application/json: schema: $ref: '#/components/schemas/Error' description: Every provider failed with a server/transport error, or inline payment settlement could not be confirmed; inspect `payment` when present before retrying. security: - bearerAuth: [] - apiKeyAuth: [] - {} summary: Google AI Overview tags: - Google x-mint: href: /api-reference/google/ai-overview x-payment-info: price: currency: USD max: '0.003360' min: '0.001800' mode: dynamic protocols: - x402: asset: '0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913' network: eip155:8453 payTo: '0x754BD2A6799e766B1E2b9f1Ae7fBdC0CeD96ef13' scheme: exact - mpp: asset: '0x20c000000000000000000000b9537d11c60e8b50' currency: USD intent: charge method: tempo network: tempo payTo: '0x754BD2A6799e766B1E2b9f1Ae7fBdC0CeD96ef13' rail: tempo scheme: exact rails: - headers: challenge: - PAYMENT-REQUIRED credential: - PAYMENT-SIGNATURE - X-PAYMENT receipt: - PAYMENT-RESPONSE - X-PAYMENT-RESPONSE pricing: currency: USD max: '0.003360' min: '0.001800' minimumUsd: '0.001000' mode: dynamic quoteBasis: floor protocol: asset: '0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913' network: eip155:8453 payTo: '0x754BD2A6799e766B1E2b9f1Ae7fBdC0CeD96ef13' scheme: exact rail: x402 settlement: onExecutionFailure: released onUnknownFinalization: indeterminate timing: after_execution - headers: challenge: - WWW-Authenticate credential: - Authorization receipt: - Payment-Receipt pricing: currency: USD max: '0.003360' min: '0.001800' minimumUsd: null mode: dynamic quoteBasis: funded_ceiling protocol: asset: '0x20c000000000000000000000b9537d11c60e8b50' currency: USD intent: charge method: tempo network: tempo payTo: '0x754BD2A6799e766B1E2b9f1Ae7fBdC0CeD96ef13' rail: tempo scheme: exact rail: mpp settlement: onExecutionFailure: charged_undelivered onVerificationError: indeterminate timing: before_execution /v1/run/google.autocomplete: post: description: 'Get Google search autocomplete suggestions for a partial query (keyword ideas). **Price:** \$0.99 per 1,000 requests (flat per request - same cost regardless of results returned). **Routing:** one lane serves this API today, so a failed attempt has nowhere to fail over to. Payment outcome follows the selected rail''s settlement policy. **Catalog:** Google Autocomplete pricing and uptime - live USD price, lane routing, and measured 30-day uptime. Every Google endpoint.' operationId: google_autocomplete parameters: - description: Optional. Comma-separated keys (dotted paths like `author.name` descend into nested objects) to keep on each result item. Keys are matched relative to each result item after the `data`/`items` envelope is unwrapped, not against the top-level response envelope, so use `jq` to reshape the whole envelope. Shrinks the response without changing cost. in: query name: fields required: false schema: type: string - description: Optional. Cap the number of result rows returned; a `_truncated` note reports how many were withheld so you can page via the API's own `limit`. Does not change cost. in: query name: max_items required: false schema: minimum: 0 type: integer - description: Optional. Return only a structural outline (top-level keys, item counts, and per-field byte sizes) instead of the full data. Does not change cost. in: query name: summary required: false schema: type: boolean - description: 'Optional. A jq expression applied to the result envelope; its output replaces `output` (multiple outputs collect into an array). Reshape freely, e.g. `jq=.data | {title, description, md: .markdown[:3500]}`. Runs sandboxed with a 250ms / 2MB budget; on failure the full result is returned with a `jqError`. Does not change cost.' in: query name: jq required: false schema: type: string - description: Optional. The most you are willing to pay for this one request, in US dollars (for example 0.05). Any route that would charge more than this is not used, so a request only ever runs on something you can afford. If nothing is available at or below your amount, the request is refused before it runs, nothing is charged, and the message tells you the cheapest price per request so you can raise it. Leave it out to accept the normal price. in: query name: max_cost_usd required: false schema: type: string - description: Optional wallet idempotency key, scoped to this customer for 24 hours. When the gateway honors the key, this synchronous in-process execution can continue after the caller disconnects, bounded by its execution deadline. A completed replayable result charges normally exactly once and can be replayed without another provider run or charge. A pending duplicate returns 409 idempotency_in_progress; reuse with different request semantics returns 409 idempotency_conflict. in: header name: Idempotency-Key required: false schema: maxLength: 255 minLength: 1 type: string requestBody: content: application/json: example: query: best coff schema: additionalProperties: false example: query: best coff properties: gl: default: us description: Two-letter country code for result localization (e.g. us, gb, de). type: string hl: default: en description: Two-letter interface and results language code for the suggestions (e.g. en, es, de). type: string preferLatencyUnderMs: description: 'Optional; omit it and routing is unchanged, with the cheapest source serving. Prefer sources whose typical response time (median over the trailing 30 days, as published on this endpoint''s lane health) is under this many milliseconds; among those, the cheapest serves. This can raise your price: when the cheapest source misses the target, a faster and dearer one serves, and you are quoted and charged its price. If no source is that fast the request is still served, by whichever source offers the best speed for its price - it is never refused for being slow. Sources we have not timed are tried last. This is a preference, not a guarantee: the median describes past requests and is not a ceiling on this one, and it excludes any wait this request itself asks for. On a paginated walk it applies to the first page only: later pages stay with the source that page chose, at the price it was quoted.' minimum: 1 type: integer query: description: The partial Google search query. type: string required: - query title: Google Autocomplete input type: object required: true responses: '200': content: application/json: schema: properties: costUsd: description: USD charged on the original run. On a replay this value is echoed for parity; the replay itself is free. type: number hint: description: 'Optional one-line nudge, absent when there is nothing to say. `large_result:` suggests the fields/max_items/summary/jq controls for a big response. `paging_unavailable:` means this result came from a source that cannot return a `nextCursor`, so it may be INCOMPLETE and cannot be continued - re-run with `requireCursor: true` to be served only by a source that can page, which may cost more per request.' type: string items: description: Number of result rows returned. For per-result SKUs the per-item cost is charged against this count; for input-priced SKUs the charge is per submitted input, independent of this count. type: integer jqError: description: Present only when a `jq` expression failed; `output` then carries the full unshaped result and this explains why the reshape did not apply. type: string output: anyOf: - additionalProperties: false properties: data: description: Result payload, or null when nothing was found. oneOf: - type: 'null' - additionalProperties: false properties: query: description: The partial query that was searched. type: string suggestions: description: Autocomplete suggestion records. Populated whenever the provider has data for the entity. items: properties: value: description: Suggested query text. Populated whenever the provider has data for the entity. type: string x-anyapi-must-populate: true required: - value type: object type: array x-anyapi-must-populate: true required: - query - suggestions type: object found: description: True when at least one autocomplete suggestion was returned. type: boolean required: - found - data title: Google Autocomplete output type: object - type: 'null' description: Normalized output, or null when the replay payload was not retained. provider: description: Always "AnyAPI". type: string replayed: description: True when this response replays the durable result of an earlier run without billing or upstream execution. type: boolean resultId: description: Opaque handle to the full unshaped result, cached ~15 min. Re-shape it for free (fields/max_items/summary/jq) via GET /v1/results/{id}, no re-billing. Absent when the result was too large to cache. type: string required: - output - provider - costUsd - items - replayed type: object description: Normalized result. headers: Idempotency-Replayed: description: Present only when the request carried an Idempotency-Key this gateway honored. "false" means the key was claimed and the run executed freshly; "true" means it replayed a prior identical run and was NOT charged again. Absent means either no key was sent or this gateway does not honor keys, so a retry could be charged twice. schema: enum: - 'false' - 'true' type: string PAYMENT-RESPONSE: description: Inline x402 payment receipt. schema: type: string Payment-Receipt: description: Inline mpp payment receipt. schema: type: string X-PAYMENT-RESPONSE: description: Inline x402 payment receipt. schema: type: string '400': content: application/json: schema: $ref: '#/components/schemas/Error' description: Input did not match the schema, or the upstream rejected the request. '401': content: application/json: schema: $ref: '#/components/schemas/Error' description: Missing or invalid API key. '402': content: application/json: schema: oneOf: - $ref: '#/components/schemas/PaymentRequiredError' - $ref: '#/components/schemas/TrialCapReached' application/problem+json: schema: properties: challengeId: minLength: 1 type: string status: const: 402 type: integer required: - status - challengeId type: object description: Insufficient wallet balance or spend cap, or an inline-payment challenge. Agent trials receive a trial_cap_reached continuation with a live RFC 8628 device authorization. headers: PAYMENT-REQUIRED: description: Inline x402 payment challenge. schema: type: string WWW-Authenticate: description: Inline mpp payment challenge. schema: type: string '404': content: application/json: schema: $ref: '#/components/schemas/Error' description: Unknown SKU, or the upstream had no matching resource. '409': content: application/json: schema: $ref: '#/components/schemas/Error' description: 'Idempotency collision. Codes: idempotency_in_progress means retry after the Retry-After delay; idempotency_conflict means use a new key for different request semantics; idempotency_needs_review means stop retrying and contact AnyAPI support with X-Anyapi-Request-Id.' '422': content: application/json: schema: $ref: '#/components/schemas/Error' description: The request was well formed but this API cannot serve the requested target site; the body's `alternatives` array lists the AnyAPI APIs that can. '429': content: application/json: schema: $ref: '#/components/schemas/Error' description: The upstream was rate limited; inspect `payment` when present before deciding whether to retry. '500': content: application/json: schema: $ref: '#/components/schemas/Error' description: An internal error prevented completion. Inspect payment when present before retrying. '502': content: application/json: schema: $ref: '#/components/schemas/Error' description: Every provider failed with a server/transport error, or inline payment settlement could not be confirmed; inspect `payment` when present before retrying. security: - bearerAuth: [] - apiKeyAuth: [] - {} summary: Google Autocomplete tags: - Google x-mint: href: /api-reference/google/autocomplete x-payment-info: price: amount: '0.000990' currency: USD mode: fixed protocols: - x402: asset: '0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913' network: eip155:8453 payTo: '0x754BD2A6799e766B1E2b9f1Ae7fBdC0CeD96ef13' scheme: exact - mpp: asset: '0x20c000000000000000000000b9537d11c60e8b50' currency: USD intent: charge method: tempo network: tempo payTo: '0x754BD2A6799e766B1E2b9f1Ae7fBdC0CeD96ef13' rail: tempo scheme: exact rails: - headers: challenge: - PAYMENT-REQUIRED credential: - PAYMENT-SIGNATURE - X-PAYMENT receipt: - PAYMENT-RESPONSE - X-PAYMENT-RESPONSE pricing: amount: '0.001000' currency: USD minimumUsd: '0.001000' mode: fixed quoteBasis: floor protocol: asset: '0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913' network: eip155:8453 payTo: '0x754BD2A6799e766B1E2b9f1Ae7fBdC0CeD96ef13' scheme: exact rail: x402 settlement: onExecutionFailure: released onUnknownFinalization: indeterminate timing: after_execution - headers: challenge: - WWW-Authenticate credential: - Authorization receipt: - Payment-Receipt pricing: amount: '0.000990' currency: USD minimumUsd: null mode: fixed quoteBasis: funded_ceiling protocol: asset: '0x20c000000000000000000000b9537d11c60e8b50' currency: USD intent: charge method: tempo network: tempo payTo: '0x754BD2A6799e766B1E2b9f1Ae7fBdC0CeD96ef13' rail: tempo scheme: exact rail: mpp settlement: onExecutionFailure: charged_undelivered onVerificationError: indeterminate timing: before_execution /v1/run/google.images: post: description: 'Run a Google Images search and get structured results: image URLs, dimensions, titles, and source pages. **Price:** billed per result - \$0.99 per 1,000 requests base + \$0.09 per 1,000 results, capped at \$1.98 per 1,000 requests. **Routing:** one lane serves this API today, so a failed attempt has nowhere to fail over to. Payment outcome follows the selected rail''s settlement policy. **Catalog:** Google Images pricing and uptime - live USD price, lane routing, and measured 30-day uptime. Every Google endpoint.' operationId: google_images parameters: - description: Optional. Comma-separated keys (dotted paths like `author.name` descend into nested objects) to keep on each result item. Keys are matched relative to each result item after the `data`/`items` envelope is unwrapped, not against the top-level response envelope, so use `jq` to reshape the whole envelope. Shrinks the response without changing cost. in: query name: fields required: false schema: type: string - description: Optional. Cap the number of result rows returned; a `_truncated` note reports how many were withheld so you can page via the API's own `limit`. Does not change cost. in: query name: max_items required: false schema: minimum: 0 type: integer - description: Optional. Return only a structural outline (top-level keys, item counts, and per-field byte sizes) instead of the full data. Does not change cost. in: query name: summary required: false schema: type: boolean - description: 'Optional. A jq expression applied to the result envelope; its output replaces `output` (multiple outputs collect into an array). Reshape freely, e.g. `jq=.data | {title, description, md: .markdown[:3500]}`. Runs sandboxed with a 250ms / 2MB budget; on failure the full result is returned with a `jqError`. Does not change cost.' in: query name: jq required: false schema: type: string - description: Optional. The most you are willing to pay for this one request, in US dollars (for example 0.05). Any route that would charge more than this is not used, so a request only ever runs on something you can afford. If nothing is available at or below your amount, the request is refused before it runs, nothing is charged, and the message tells you the cheapest price per request so you can raise it. Leave it out to accept the normal price. in: query name: max_cost_usd required: false schema: type: string - description: Optional wallet idempotency key, scoped to this customer for 24 hours. When the gateway honors the key, this synchronous in-process execution can continue after the caller disconnects, bounded by its execution deadline. A completed replayable result charges normally exactly once and can be replayed without another provider run or charge. A pending duplicate returns 409 idempotency_in_progress; reuse with different request semantics returns 409 idempotency_conflict. in: header name: Idempotency-Key required: false schema: maxLength: 255 minLength: 1 type: string requestBody: content: application/json: example: gl: us hl: en limit: 5 query: golden retriever schema: additionalProperties: false example: gl: us hl: en limit: 5 query: golden retriever properties: autocorrect: description: Toggle Google spelling autocorrect (default true). Set false to search the exact query without correction. type: boolean gl: default: us description: Two-letter country code for result localization (e.g. us, gb, de). type: string hl: default: en description: Two-letter interface and results language code (e.g. en, es, de). type: string limit: default: 20 description: Maximum number of images to return (1-100, default 20). Requests for 10 results or fewer are billed at a lower rate than larger requests. maximum: 100 minimum: 1 type: integer location: description: Fine-grained location for result localization, given as a canonical Google location string (e.g. 'New York, United States', 'London, United Kingdom'). More specific than the country-level gl. type: string preferLatencyUnderMs: description: 'Optional; omit it and routing is unchanged, with the cheapest source serving. Prefer sources whose typical response time (median over the trailing 30 days, as published on this endpoint''s lane health) is under this many milliseconds; among those, the cheapest serves. This can raise your price: when the cheapest source misses the target, a faster and dearer one serves, and you are quoted and charged its price. If no source is that fast the request is still served, by whichever source offers the best speed for its price - it is never refused for being slow. Sources we have not timed are tried last. This is a preference, not a guarantee: the median describes past requests and is not a ceiling on this one, and it excludes any wait this request itself asks for. On a paginated walk it applies to the first page only: later pages stay with the source that page chose, at the price it was quoted.' minimum: 1 type: integer query: description: Image search query (e.g. golden gate bridge at sunset). type: string timeframe: description: 'Restrict results to a recent time window: 1h, 1d, 7d, 1y, or all. Default all (no time restriction).' type: string required: - query title: Google Images input type: object required: true responses: '200': content: application/json: schema: properties: costUsd: description: USD charged on the original run. On a replay this value is echoed for parity; the replay itself is free. type: number hint: description: 'Optional one-line nudge, absent when there is nothing to say. `large_result:` suggests the fields/max_items/summary/jq controls for a big response. `paging_unavailable:` means this result came from a source that cannot return a `nextCursor`, so it may be INCOMPLETE and cannot be continued - re-run with `requireCursor: true` to be served only by a source that can page, which may cost more per request.' type: string items: description: Number of result rows returned. For per-result SKUs the per-item cost is charged against this count; for input-priced SKUs the charge is per submitted input, independent of this count. type: integer jqError: description: Present only when a `jq` expression failed; `output` then carries the full unshaped result and this explains why the reshape did not apply. type: string output: anyOf: - additionalProperties: false properties: data: description: The result wrapper, or null when nothing was found. oneOf: - type: 'null' - additionalProperties: false properties: items: description: 'Image result records: image URL, dimensions, title, and the source page it appears on. Populated whenever the provider has data for the entity.' items: properties: height: description: Full image height in pixels. type: integer source: description: Host domain of the page the image appears on. Populated whenever the provider has data for the entity. type: string x-anyapi-must-populate: true sourceUrl: description: URL of the page the image appears on. Populated whenever the provider has data for the entity. type: string x-anyapi-must-populate: true thumbnailUrl: description: URL to a thumbnail of the image. type: string title: description: Image result title. Populated whenever the provider has data for the entity. type: string x-anyapi-must-populate: true url: description: Direct URL to the full-size image. Populated whenever the provider has data for the entity. type: string x-anyapi-must-populate: true width: description: Full image width in pixels. type: integer required: - url - title type: object type: array x-anyapi-must-populate: true required: - items type: object found: description: True when at least one image result was returned. type: boolean required: - found - data title: Google Images output type: object - type: 'null' description: Normalized output, or null when the replay payload was not retained. provider: description: Always "AnyAPI". type: string replayed: description: True when this response replays the durable result of an earlier run without billing or upstream execution. type: boolean resultId: description: Opaque handle to the full unshaped result, cached ~15 min. Re-shape it for free (fields/max_items/summary/jq) via GET /v1/results/{id}, no re-billing. Absent when the result was too large to cache. type: string required: - output - provider - costUsd - items - replayed type: object description: Normalized result. headers: Idempotency-Replayed: description: Present only when the request carried an Idempotency-Key this gateway honored. "false" means the key was claimed and the run executed freshly; "true" means it replayed a prior identical run and was NOT charged again. Absent means either no key was sent or this gateway does not honor keys, so a retry could be charged twice. schema: enum: - 'false' - 'true' type: string PAYMENT-RESPONSE: description: Inline x402 payment receipt. schema: type: string Payment-Receipt: description: Inline mpp payment receipt. schema: type: string X-PAYMENT-RESPONSE: description: Inline x402 payment receipt. schema: type: string '400': content: application/json: schema: $ref: '#/components/schemas/Error' description: Input did not match the schema, or the upstream rejected the request. '401': content: application/json: schema: $ref: '#/components/schemas/Error' description: Missing or invalid API key. '402': content: application/json: schema: oneOf: - $ref: '#/components/schemas/PaymentRequiredError' - $ref: '#/components/schemas/TrialCapReached' application/problem+json: schema: properties: challengeId: minLength: 1 type: string status: const: 402 type: integer required: - status - challengeId type: object description: Insufficient wallet balance or spend cap, or an inline-payment challenge. Agent trials receive a trial_cap_reached continuation with a live RFC 8628 device authorization. headers: PAYMENT-REQUIRED: description: Inline x402 payment challenge. schema: type: string WWW-Authenticate: description: Inline mpp payment challenge. schema: type: string '404': content: application/json: schema: $ref: '#/components/schemas/Error' description: Unknown SKU, or the upstream had no matching resource. '409': content: application/json: schema: $ref: '#/components/schemas/Error' description: 'Idempotency collision. Codes: idempotency_in_progress means retry after the Retry-After delay; idempotency_conflict means use a new key for different request semantics; idempotency_needs_review means stop retrying and contact AnyAPI support with X-Anyapi-Request-Id.' '422': content: application/json: schema: $ref: '#/components/schemas/Error' description: The request was well formed but this API cannot serve the requested target site; the body's `alternatives` array lists the AnyAPI APIs that can. '429': content: application/json: schema: $ref: '#/components/schemas/Error' description: The upstream was rate limited; inspect `payment` when present before deciding whether to retry. '500': content: application/json: schema: $ref: '#/components/schemas/Error' description: An internal error prevented completion. Inspect payment when present before retrying. '502': content: application/json: schema: $ref: '#/components/schemas/Error' description: Every provider failed with a server/transport error, or inline payment settlement could not be confirmed; inspect `payment` when present before retrying. security: - bearerAuth: [] - apiKeyAuth: [] - {} summary: Google Images tags: - Google x-mint: href: /api-reference/google/images x-payment-info: price: currency: USD max: '0.001980' min: '0.000990' mode: dynamic protocols: - x402: asset: '0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913' network: eip155:8453 payTo: '0x754BD2A6799e766B1E2b9f1Ae7fBdC0CeD96ef13' scheme: exact - mpp: asset: '0x20c000000000000000000000b9537d11c60e8b50' currency: USD intent: charge method: tempo network: tempo payTo: '0x754BD2A6799e766B1E2b9f1Ae7fBdC0CeD96ef13' rail: tempo scheme: exact rails: - headers: challenge: - PAYMENT-REQUIRED credential: - PAYMENT-SIGNATURE - X-PAYMENT receipt: - PAYMENT-RESPONSE - X-PAYMENT-RESPONSE pricing: currency: USD max: '0.001980' min: '0.001000' minimumUsd: '0.001000' mode: dynamic quoteBasis: floor protocol: asset: '0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913' network: eip155:8453 payTo: '0x754BD2A6799e766B1E2b9f1Ae7fBdC0CeD96ef13' scheme: exact rail: x402 settlement: onExecutionFailure: released onUnknownFinalization: indeterminate timing: after_execution - headers: challenge: - WWW-Authenticate credential: - Authorization receipt: - Payment-Receipt pricing: currency: USD max: '0.001980' min: '0.000990' minimumUsd: null mode: dynamic quoteBasis: funded_ceiling protocol: asset: '0x20c000000000000000000000b9537d11c60e8b50' currency: USD intent: charge method: tempo network: tempo payTo: '0x754BD2A6799e766B1E2b9f1Ae7fBdC0CeD96ef13' rail: tempo scheme: exact rail: mpp settlement: onExecutionFailure: charged_undelivered onVerificationError: indeterminate timing: before_execution /v1/run/google.lens: post: description: 'Reverse image search: find web pages and visual matches for an image URL. **Price:** \$2.97 per 1,000 requests (flat per request - same cost regardless of results returned). **Routing:** one lane serves this API today, so a failed attempt has nowhere to fail over to. Payment outcome follows the selected rail''s settlement policy. **Catalog:** Google Lens pricing and uptime - live USD price, lane routing, and measured 30-day uptime. Every Google endpoint.' operationId: google_lens parameters: - description: Optional. Comma-separated keys (dotted paths like `author.name` descend into nested objects) to keep on each result item. Keys are matched relative to each result item after the `data`/`items` envelope is unwrapped, not against the top-level response envelope, so use `jq` to reshape the whole envelope. Shrinks the response without changing cost. in: query name: fields required: false schema: type: string - description: Optional. Cap the number of result rows returned; a `_truncated` note reports how many were withheld so you can page via the API's own `limit`. Does not change cost. in: query name: max_items required: false schema: minimum: 0 type: integer - description: Optional. Return only a structural outline (top-level keys, item counts, and per-field byte sizes) instead of the full data. Does not change cost. in: query name: summary required: false schema: type: boolean - description: 'Optional. A jq expression applied to the result envelope; its output replaces `output` (multiple outputs collect into an array). Reshape freely, e.g. `jq=.data | {title, description, md: .markdown[:3500]}`. Runs sandboxed with a 250ms / 2MB budget; on failure the full result is returned with a `jqError`. Does not change cost.' in: query name: jq required: false schema: type: string - description: Optional. The most you are willing to pay for this one request, in US dollars (for example 0.05). Any route that would charge more than this is not used, so a request only ever runs on something you can afford. If nothing is available at or below your amount, the request is refused before it runs, nothing is charged, and the message tells you the cheapest price per request so you can raise it. Leave it out to accept the normal price. in: query name: max_cost_usd required: false schema: type: string - description: Optional wallet idempotency key, scoped to this customer for 24 hours. When the gateway honors the key, this synchronous in-process execution can continue after the caller disconnects, bounded by its execution deadline. A completed replayable result charges normally exactly once and can be replayed without another provider run or charge. A pending duplicate returns 409 idempotency_in_progress; reuse with different request semantics returns 409 idempotency_conflict. in: header name: Idempotency-Key required: false schema: maxLength: 255 minLength: 1 type: string requestBody: content: application/json: example: url: https://i.imgur.com/HBrB8p0.png schema: additionalProperties: false example: url: https://i.imgur.com/HBrB8p0.png properties: preferLatencyUnderMs: description: 'Optional; omit it and routing is unchanged, with the cheapest source serving. Prefer sources whose typical response time (median over the trailing 30 days, as published on this endpoint''s lane health) is under this many milliseconds; among those, the cheapest serves. This can raise your price: when the cheapest source misses the target, a faster and dearer one serves, and you are quoted and charged its price. If no source is that fast the request is still served, by whichever source offers the best speed for its price - it is never refused for being slow. Sources we have not timed are tried last. This is a preference, not a guarantee: the median describes past requests and is not a ceiling on this one, and it excludes any wait this request itself asks for. On a paginated walk it applies to the first page only: later pages stay with the source that page chose, at the price it was quoted.' minimum: 1 type: integer url: description: Public URL of the image to search with. format: uri type: string required: - url title: Google Lens input type: object required: true responses: '200': content: application/json: schema: properties: costUsd: description: USD charged on the original run. On a replay this value is echoed for parity; the replay itself is free. type: number hint: description: 'Optional one-line nudge, absent when there is nothing to say. `large_result:` suggests the fields/max_items/summary/jq controls for a big response. `paging_unavailable:` means this result came from a source that cannot return a `nextCursor`, so it may be INCOMPLETE and cannot be continued - re-run with `requireCursor: true` to be served only by a source that can page, which may cost more per request.' type: string items: description: Number of result rows returned. For per-result SKUs the per-item cost is charged against this count; for input-priced SKUs the charge is per submitted input, independent of this count. type: integer jqError: description: Present only when a `jq` expression failed; `output` then carries the full unshaped result and this explains why the reshape did not apply. type: string output: anyOf: - additionalProperties: false properties: data: description: Result payload, or null when nothing was found. oneOf: - type: 'null' - additionalProperties: false properties: results: description: Visual match result records. Populated whenever the provider has data for the entity. items: properties: image: description: Matched image URL. format: uri type: string link: description: URL to the matching web page. Populated whenever the provider has data for the entity. format: uri type: string x-anyapi-must-populate: true source: description: Source site name. type: string thumbnailUrl: description: Thumbnail image URL for the match. format: uri type: string title: description: Title of the matching web page. Populated whenever the provider has data for the entity. type: string x-anyapi-must-populate: true required: - title - link type: object type: array x-anyapi-must-populate: true url: description: The input image URL that was searched. type: string required: - url - results type: object found: description: True when at least one visual match was returned. type: boolean required: - found - data title: Google Lens output type: object - type: 'null' description: Normalized output, or null when the replay payload was not retained. provider: description: Always "AnyAPI". type: string replayed: description: True when this response replays the durable result of an earlier run without billing or upstream execution. type: boolean resultId: description: Opaque handle to the full unshaped result, cached ~15 min. Re-shape it for free (fields/max_items/summary/jq) via GET /v1/results/{id}, no re-billing. Absent when the result was too large to cache. type: string required: - output - provider - costUsd - items - replayed type: object description: Normalized result. headers: Idempotency-Replayed: description: Present only when the request carried an Idempotency-Key this gateway honored. "false" means the key was claimed and the run executed freshly; "true" means it replayed a prior identical run and was NOT charged again. Absent means either no key was sent or this gateway does not honor keys, so a retry could be charged twice. schema: enum: - 'false' - 'true' type: string PAYMENT-RESPONSE: description: Inline x402 payment receipt. schema: type: string Payment-Receipt: description: Inline mpp payment receipt. schema: type: string X-PAYMENT-RESPONSE: description: Inline x402 payment receipt. schema: type: string '400': content: application/json: schema: $ref: '#/components/schemas/Error' description: Input did not match the schema, or the upstream rejected the request. '401': content: application/json: schema: $ref: '#/components/schemas/Error' description: Missing or invalid API key. '402': content: application/json: schema: oneOf: - $ref: '#/components/schemas/PaymentRequiredError' - $ref: '#/components/schemas/TrialCapReached' application/problem+json: schema: properties: challengeId: minLength: 1 type: string status: const: 402 type: integer required: - status - challengeId type: object description: Insufficient wallet balance or spend cap, or an inline-payment challenge. Agent trials receive a trial_cap_reached continuation with a live RFC 8628 device authorization. headers: PAYMENT-REQUIRED: description: Inline x402 payment challenge. schema: type: string WWW-Authenticate: description: Inline mpp payment challenge. schema: type: string '404': content: application/json: schema: $ref: '#/components/schemas/Error' description: Unknown SKU, or the upstream had no matching resource. '409': content: application/json: schema: $ref: '#/components/schemas/Error' description: 'Idempotency collision. Codes: idempotency_in_progress means retry after the Retry-After delay; idempotency_conflict means use a new key for different request semantics; idempotency_needs_review means stop retrying and contact AnyAPI support with X-Anyapi-Request-Id.' '422': content: application/json: schema: $ref: '#/components/schemas/Error' description: The request was well formed but this API cannot serve the requested target site; the body's `alternatives` array lists the AnyAPI APIs that can. '429': content: application/json: schema: $ref: '#/components/schemas/Error' description: The upstream was rate limited; inspect `payment` when present before deciding whether to retry. '500': content: application/json: schema: $ref: '#/components/schemas/Error' description: An internal error prevented completion. Inspect payment when present before retrying. '502': content: application/json: schema: $ref: '#/components/schemas/Error' description: Every provider failed with a server/transport error, or inline payment settlement could not be confirmed; inspect `payment` when present before retrying. security: - bearerAuth: [] - apiKeyAuth: [] - {} summary: Google Lens tags: - Google x-mint: href: /api-reference/google/lens x-payment-info: price: amount: '0.002970' currency: USD mode: fixed protocols: - x402: asset: '0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913' network: eip155:8453 payTo: '0x754BD2A6799e766B1E2b9f1Ae7fBdC0CeD96ef13' scheme: exact - mpp: asset: '0x20c000000000000000000000b9537d11c60e8b50' currency: USD intent: charge method: tempo network: tempo payTo: '0x754BD2A6799e766B1E2b9f1Ae7fBdC0CeD96ef13' rail: tempo scheme: exact rails: - headers: challenge: - PAYMENT-REQUIRED credential: - PAYMENT-SIGNATURE - X-PAYMENT receipt: - PAYMENT-RESPONSE - X-PAYMENT-RESPONSE pricing: amount: '0.002970' currency: USD minimumUsd: '0.001000' mode: fixed quoteBasis: floor protocol: asset: '0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913' network: eip155:8453 payTo: '0x754BD2A6799e766B1E2b9f1Ae7fBdC0CeD96ef13' scheme: exact rail: x402 settlement: onExecutionFailure: released onUnknownFinalization: indeterminate timing: after_execution - headers: challenge: - WWW-Authenticate credential: - Authorization receipt: - Payment-Receipt pricing: amount: '0.002970' currency: USD minimumUsd: null mode: fixed quoteBasis: funded_ceiling protocol: asset: '0x20c000000000000000000000b9537d11c60e8b50' currency: USD intent: charge method: tempo network: tempo payTo: '0x754BD2A6799e766B1E2b9f1Ae7fBdC0CeD96ef13' rail: tempo scheme: exact rail: mpp settlement: onExecutionFailure: charged_undelivered onVerificationError: indeterminate timing: before_execution /v1/run/google.news: post: description: 'Search Google News by keyword and get fresh articles (headlines, sources, links, and publish times) as clean JSON. **Price:** \$0.99 per 1,000 requests (flat per request - same cost regardless of results returned). **Routing:** one lane serves this API today, so a failed attempt has nowhere to fail over to. Payment outcome follows the selected rail''s settlement policy. **Catalog:** Google News pricing and uptime - live USD price, lane routing, and measured 30-day uptime. Every Google endpoint.' operationId: google_news parameters: - description: Optional. Comma-separated keys (dotted paths like `author.name` descend into nested objects) to keep on each result item. Keys are matched relative to each result item after the `data`/`items` envelope is unwrapped, not against the top-level response envelope, so use `jq` to reshape the whole envelope. Shrinks the response without changing cost. in: query name: fields required: false schema: type: string - description: Optional. Cap the number of result rows returned; a `_truncated` note reports how many were withheld so you can page via the API's own `limit`. Does not change cost. in: query name: max_items required: false schema: minimum: 0 type: integer - description: Optional. Return only a structural outline (top-level keys, item counts, and per-field byte sizes) instead of the full data. Does not change cost. in: query name: summary required: false schema: type: boolean - description: 'Optional. A jq expression applied to the result envelope; its output replaces `output` (multiple outputs collect into an array). Reshape freely, e.g. `jq=.data | {title, description, md: .markdown[:3500]}`. Runs sandboxed with a 250ms / 2MB budget; on failure the full result is returned with a `jqError`. Does not change cost.' in: query name: jq required: false schema: type: string - description: Optional. The most you are willing to pay for this one request, in US dollars (for example 0.05). Any route that would charge more than this is not used, so a request only ever runs on something you can afford. If nothing is available at or below your amount, the request is refused before it runs, nothing is charged, and the message tells you the cheapest price per request so you can raise it. Leave it out to accept the normal price. in: query name: max_cost_usd required: false schema: type: string - description: Optional wallet idempotency key, scoped to this customer for 24 hours. When the gateway honors the key, this synchronous in-process execution can continue after the caller disconnects, bounded by its execution deadline. A completed replayable result charges normally exactly once and can be replayed without another provider run or charge. A pending duplicate returns 409 idempotency_in_progress; reuse with different request semantics returns 409 idempotency_conflict. in: header name: Idempotency-Key required: false schema: maxLength: 255 minLength: 1 type: string requestBody: content: application/json: example: gl: us hl: en query: openai schema: additionalProperties: false example: gl: us hl: en query: openai properties: gl: default: us description: Two-letter country code for result localization (e.g. us, gb, de). type: string hl: default: en description: Two-letter interface and results language code (e.g. en, es, de). type: string limit: description: Requested article count (1-20, default 20). Google News returns its latest matching articles and may return more or fewer than requested. Price is flat per request. maximum: 20 minimum: 1 type: integer location: description: Fine-grained location for result localization, given as a canonical Google location string (e.g. 'New York, United States', 'London, United Kingdom'). More specific than the country-level gl. type: string preferLatencyUnderMs: description: 'Optional; omit it and routing is unchanged, with the cheapest source serving. Prefer sources whose typical response time (median over the trailing 30 days, as published on this endpoint''s lane health) is under this many milliseconds; among those, the cheapest serves. This can raise your price: when the cheapest source misses the target, a faster and dearer one serves, and you are quoted and charged its price. If no source is that fast the request is still served, by whichever source offers the best speed for its price - it is never refused for being slow. Sources we have not timed are tried last. This is a preference, not a guarantee: the median describes past requests and is not a ceiling on this one, and it excludes any wait this request itself asks for. On a paginated walk it applies to the first page only: later pages stay with the source that page chose, at the price it was quoted.' minimum: 1 type: integer query: description: News search query; supports operators like '-', 'OR', and 'site:' (e.g. bitcoin site:cnn.com). type: string timeframe: description: 'How far back to search, written as a count plus a unit: h hours, d days, w weeks, m months, y years. So 2h is the last two hours, 30d the last thirty days, 6m the last six months. Omit this field, or send ''all'', to search without a time limit.' pattern: ^(all|[1-9][0-9]*[hdwmy])$ type: string required: - query title: Google News input type: object required: true responses: '200': content: application/json: schema: properties: costUsd: description: USD charged on the original run. On a replay this value is echoed for parity; the replay itself is free. type: number hint: description: 'Optional one-line nudge, absent when there is nothing to say. `large_result:` suggests the fields/max_items/summary/jq controls for a big response. `paging_unavailable:` means this result came from a source that cannot return a `nextCursor`, so it may be INCOMPLETE and cannot be continued - re-run with `requireCursor: true` to be served only by a source that can page, which may cost more per request.' type: string items: description: Number of result rows returned. For per-result SKUs the per-item cost is charged against this count; for input-priced SKUs the charge is per submitted input, independent of this count. type: integer jqError: description: Present only when a `jq` expression failed; `output` then carries the full unshaped result and this explains why the reshape did not apply. type: string output: anyOf: - additionalProperties: false properties: data: description: Result payload, or null when nothing was found. oneOf: - type: 'null' - additionalProperties: false properties: items: description: 'Article records: headline, source name, article link, and publish time. Populated whenever the provider has data for the entity.' items: properties: createdUtc: description: UTC epoch timestamp in seconds (Unix time). Multiply by 1000 for a JS Date in milliseconds. Populated whenever the provider has data for the entity. type: number x-anyapi-must-populate: true snippet: description: Article snippet when available. type: string source: description: Publisher name. Populated whenever the provider has data for the entity. type: string x-anyapi-must-populate: true title: description: Article headline. Populated whenever the provider has data for the entity. type: string x-anyapi-must-populate: true url: description: Article link. Populated whenever the provider has data for the entity. type: string x-anyapi-must-populate: true required: - title - url type: object type: array x-anyapi-must-populate: true required: - items type: object found: description: True when the search returned at least one article. type: boolean required: - found - data title: Google News output type: object - type: 'null' description: Normalized output, or null when the replay payload was not retained. provider: description: Always "AnyAPI". type: string replayed: description: True when this response replays the durable result of an earlier run without billing or upstream execution. type: boolean resultId: description: Opaque handle to the full unshaped result, cached ~15 min. Re-shape it for free (fields/max_items/summary/jq) via GET /v1/results/{id}, no re-billing. Absent when the result was too large to cache. type: string required: - output - provider - costUsd - items - replayed type: object description: Normalized result. headers: Idempotency-Replayed: description: Present only when the request carried an Idempotency-Key this gateway honored. "false" means the key was claimed and the run executed freshly; "true" means it replayed a prior identical run and was NOT charged again. Absent means either no key was sent or this gateway does not honor keys, so a retry could be charged twice. schema: enum: - 'false' - 'true' type: string PAYMENT-RESPONSE: description: Inline x402 payment receipt. schema: type: string Payment-Receipt: description: Inline mpp payment receipt. schema: type: string X-PAYMENT-RESPONSE: description: Inline x402 payment receipt. schema: type: string '400': content: application/json: schema: $ref: '#/components/schemas/Error' description: Input did not match the schema, or the upstream rejected the request. '401': content: application/json: schema: $ref: '#/components/schemas/Error' description: Missing or invalid API key. '402': content: application/json: schema: oneOf: - $ref: '#/components/schemas/PaymentRequiredError' - $ref: '#/components/schemas/TrialCapReached' application/problem+json: schema: properties: challengeId: minLength: 1 type: string status: const: 402 type: integer required: - status - challengeId type: object description: Insufficient wallet balance or spend cap, or an inline-payment challenge. Agent trials receive a trial_cap_reached continuation with a live RFC 8628 device authorization. headers: PAYMENT-REQUIRED: description: Inline x402 payment challenge. schema: type: string WWW-Authenticate: description: Inline mpp payment challenge. schema: type: string '404': content: application/json: schema: $ref: '#/components/schemas/Error' description: Unknown SKU, or the upstream had no matching resource. '409': content: application/json: schema: $ref: '#/components/schemas/Error' description: 'Idempotency collision. Codes: idempotency_in_progress means retry after the Retry-After delay; idempotency_conflict means use a new key for different request semantics; idempotency_needs_review means stop retrying and contact AnyAPI support with X-Anyapi-Request-Id.' '422': content: application/json: schema: $ref: '#/components/schemas/Error' description: The request was well formed but this API cannot serve the requested target site; the body's `alternatives` array lists the AnyAPI APIs that can. '429': content: application/json: schema: $ref: '#/components/schemas/Error' description: The upstream was rate limited; inspect `payment` when present before deciding whether to retry. '500': content: application/json: schema: $ref: '#/components/schemas/Error' description: An internal error prevented completion. Inspect payment when present before retrying. '502': content: application/json: schema: $ref: '#/components/schemas/Error' description: Every provider failed with a server/transport error, or inline payment settlement could not be confirmed; inspect `payment` when present before retrying. security: - bearerAuth: [] - apiKeyAuth: [] - {} summary: Google News tags: - Google x-mint: href: /api-reference/google/news x-payment-info: price: amount: '0.000990' currency: USD mode: fixed protocols: - x402: asset: '0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913' network: eip155:8453 payTo: '0x754BD2A6799e766B1E2b9f1Ae7fBdC0CeD96ef13' scheme: exact - mpp: asset: '0x20c000000000000000000000b9537d11c60e8b50' currency: USD intent: charge method: tempo network: tempo payTo: '0x754BD2A6799e766B1E2b9f1Ae7fBdC0CeD96ef13' rail: tempo scheme: exact rails: - headers: challenge: - PAYMENT-REQUIRED credential: - PAYMENT-SIGNATURE - X-PAYMENT receipt: - PAYMENT-RESPONSE - X-PAYMENT-RESPONSE pricing: amount: '0.001000' currency: USD minimumUsd: '0.001000' mode: fixed quoteBasis: floor protocol: asset: '0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913' network: eip155:8453 payTo: '0x754BD2A6799e766B1E2b9f1Ae7fBdC0CeD96ef13' scheme: exact rail: x402 settlement: onExecutionFailure: released onUnknownFinalization: indeterminate timing: after_execution - headers: challenge: - WWW-Authenticate credential: - Authorization receipt: - Payment-Receipt pricing: amount: '0.000990' currency: USD minimumUsd: null mode: fixed quoteBasis: funded_ceiling protocol: asset: '0x20c000000000000000000000b9537d11c60e8b50' currency: USD intent: charge method: tempo network: tempo payTo: '0x754BD2A6799e766B1E2b9f1Ae7fBdC0CeD96ef13' rail: tempo scheme: exact rail: mpp settlement: onExecutionFailure: charged_undelivered onVerificationError: indeterminate timing: before_execution /v1/run/google.patents: post: description: 'Search Google Patents with title, patent number, inventor, assignee, key dates, and PDF link. **Price:** \$0.99 per 1,000 requests (flat per request - same cost regardless of results returned). **Routing:** one lane serves this API today, so a failed attempt has nowhere to fail over to. Payment outcome follows the selected rail''s settlement policy. **Catalog:** Google Patents pricing and uptime - live USD price, lane routing, and measured 30-day uptime. Every Google endpoint.' operationId: google_patents parameters: - description: Optional. Comma-separated keys (dotted paths like `author.name` descend into nested objects) to keep on each result item. Keys are matched relative to each result item after the `data`/`items` envelope is unwrapped, not against the top-level response envelope, so use `jq` to reshape the whole envelope. Shrinks the response without changing cost. in: query name: fields required: false schema: type: string - description: Optional. Cap the number of result rows returned; a `_truncated` note reports how many were withheld so you can page via the API's own `limit`. Does not change cost. in: query name: max_items required: false schema: minimum: 0 type: integer - description: Optional. Return only a structural outline (top-level keys, item counts, and per-field byte sizes) instead of the full data. Does not change cost. in: query name: summary required: false schema: type: boolean - description: 'Optional. A jq expression applied to the result envelope; its output replaces `output` (multiple outputs collect into an array). Reshape freely, e.g. `jq=.data | {title, description, md: .markdown[:3500]}`. Runs sandboxed with a 250ms / 2MB budget; on failure the full result is returned with a `jqError`. Does not change cost.' in: query name: jq required: false schema: type: string - description: Optional. The most you are willing to pay for this one request, in US dollars (for example 0.05). Any route that would charge more than this is not used, so a request only ever runs on something you can afford. If nothing is available at or below your amount, the request is refused before it runs, nothing is charged, and the message tells you the cheapest price per request so you can raise it. Leave it out to accept the normal price. in: query name: max_cost_usd required: false schema: type: string - description: Optional wallet idempotency key, scoped to this customer for 24 hours. When the gateway honors the key, this synchronous in-process execution can continue after the caller disconnects, bounded by its execution deadline. A completed replayable result charges normally exactly once and can be replayed without another provider run or charge. A pending duplicate returns 409 idempotency_in_progress; reuse with different request semantics returns 409 idempotency_conflict. in: header name: Idempotency-Key required: false schema: maxLength: 255 minLength: 1 type: string requestBody: content: application/json: example: query: wireless charging schema: additionalProperties: false example: query: wireless charging properties: preferLatencyUnderMs: description: 'Optional; omit it and routing is unchanged, with the cheapest source serving. Prefer sources whose typical response time (median over the trailing 30 days, as published on this endpoint''s lane health) is under this many milliseconds; among those, the cheapest serves. This can raise your price: when the cheapest source misses the target, a faster and dearer one serves, and you are quoted and charged its price. If no source is that fast the request is still served, by whichever source offers the best speed for its price - it is never refused for being slow. Sources we have not timed are tried last. This is a preference, not a guarantee: the median describes past requests and is not a ceiling on this one, and it excludes any wait this request itself asks for. On a paginated walk it applies to the first page only: later pages stay with the source that page chose, at the price it was quoted.' minimum: 1 type: integer query: description: The Google Patents search query. type: string required: - query title: Google Patents input type: object required: true responses: '200': content: application/json: schema: properties: costUsd: description: USD charged on the original run. On a replay this value is echoed for parity; the replay itself is free. type: number hint: description: 'Optional one-line nudge, absent when there is nothing to say. `large_result:` suggests the fields/max_items/summary/jq controls for a big response. `paging_unavailable:` means this result came from a source that cannot return a `nextCursor`, so it may be INCOMPLETE and cannot be continued - re-run with `requireCursor: true` to be served only by a source that can page, which may cost more per request.' type: string items: description: Number of result rows returned. For per-result SKUs the per-item cost is charged against this count; for input-priced SKUs the charge is per submitted input, independent of this count. type: integer jqError: description: Present only when a `jq` expression failed; `output` then carries the full unshaped result and this explains why the reshape did not apply. type: string output: anyOf: - additionalProperties: false properties: data: description: Result payload, or null when nothing was found. oneOf: - type: 'null' - additionalProperties: false properties: query: description: The query that was searched. type: string results: description: Patent result records. items: properties: assignee: description: Patent assignee. type: string filedUtc: description: UTC epoch timestamp in seconds (Unix time). Multiply by 1000 for a JS Date in milliseconds. type: number grantedUtc: description: UTC epoch timestamp in seconds (Unix time). Multiply by 1000 for a JS Date in milliseconds. type: number image: description: First patent figure thumbnail image URL. format: uri type: string inventor: description: Named inventor or inventors. type: string link: description: URL to the patent. Populated whenever the provider has data for the entity. format: uri type: string x-anyapi-must-populate: true pdfUrl: description: URL to an available patent PDF. format: uri type: string priorityUtc: description: UTC epoch timestamp in seconds (Unix time). Multiply by 1000 for a JS Date in milliseconds. type: number publicationNumber: description: Patent publication number (e.g. US11303135B2). Populated whenever the provider has data for the entity. type: string x-anyapi-must-populate: true publishedUtc: description: UTC epoch timestamp in seconds (Unix time). Multiply by 1000 for a JS Date in milliseconds. type: number snippet: description: Short patent description snippet. type: string title: description: Patent title. Populated whenever the provider has data for the entity. type: string x-anyapi-must-populate: true required: - title - link - publicationNumber type: object type: array required: - query - results type: object found: description: True when at least one patent result was returned. type: boolean required: - found - data title: Google Patents output type: object - type: 'null' description: Normalized output, or null when the replay payload was not retained. provider: description: Always "AnyAPI". type: string replayed: description: True when this response replays the durable result of an earlier run without billing or upstream execution. type: boolean resultId: description: Opaque handle to the full unshaped result, cached ~15 min. Re-shape it for free (fields/max_items/summary/jq) via GET /v1/results/{id}, no re-billing. Absent when the result was too large to cache. type: string required: - output - provider - costUsd - items - replayed type: object description: Normalized result. headers: Idempotency-Replayed: description: Present only when the request carried an Idempotency-Key this gateway honored. "false" means the key was claimed and the run executed freshly; "true" means it replayed a prior identical run and was NOT charged again. Absent means either no key was sent or this gateway does not honor keys, so a retry could be charged twice. schema: enum: - 'false' - 'true' type: string PAYMENT-RESPONSE: description: Inline x402 payment receipt. schema: type: string Payment-Receipt: description: Inline mpp payment receipt. schema: type: string X-PAYMENT-RESPONSE: description: Inline x402 payment receipt. schema: type: string '400': content: application/json: schema: $ref: '#/components/schemas/Error' description: Input did not match the schema, or the upstream rejected the request. '401': content: application/json: schema: $ref: '#/components/schemas/Error' description: Missing or invalid API key. '402': content: application/json: schema: oneOf: - $ref: '#/components/schemas/PaymentRequiredError' - $ref: '#/components/schemas/TrialCapReached' application/problem+json: schema: properties: challengeId: minLength: 1 type: string status: const: 402 type: integer required: - status - challengeId type: object description: Insufficient wallet balance or spend cap, or an inline-payment challenge. Agent trials receive a trial_cap_reached continuation with a live RFC 8628 device authorization. headers: PAYMENT-REQUIRED: description: Inline x402 payment challenge. schema: type: string WWW-Authenticate: description: Inline mpp payment challenge. schema: type: string '404': content: application/json: schema: $ref: '#/components/schemas/Error' description: Unknown SKU, or the upstream had no matching resource. '409': content: application/json: schema: $ref: '#/components/schemas/Error' description: 'Idempotency collision. Codes: idempotency_in_progress means retry after the Retry-After delay; idempotency_conflict means use a new key for different request semantics; idempotency_needs_review means stop retrying and contact AnyAPI support with X-Anyapi-Request-Id.' '422': content: application/json: schema: $ref: '#/components/schemas/Error' description: The request was well formed but this API cannot serve the requested target site; the body's `alternatives` array lists the AnyAPI APIs that can. '429': content: application/json: schema: $ref: '#/components/schemas/Error' description: The upstream was rate limited; inspect `payment` when present before deciding whether to retry. '500': content: application/json: schema: $ref: '#/components/schemas/Error' description: An internal error prevented completion. Inspect payment when present before retrying. '502': content: application/json: schema: $ref: '#/components/schemas/Error' description: Every provider failed with a server/transport error, or inline payment settlement could not be confirmed; inspect `payment` when present before retrying. security: - bearerAuth: [] - apiKeyAuth: [] - {} summary: Google Patents tags: - Google x-mint: href: /api-reference/google/patents x-payment-info: price: amount: '0.000990' currency: USD mode: fixed protocols: - x402: asset: '0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913' network: eip155:8453 payTo: '0x754BD2A6799e766B1E2b9f1Ae7fBdC0CeD96ef13' scheme: exact - mpp: asset: '0x20c000000000000000000000b9537d11c60e8b50' currency: USD intent: charge method: tempo network: tempo payTo: '0x754BD2A6799e766B1E2b9f1Ae7fBdC0CeD96ef13' rail: tempo scheme: exact rails: - headers: challenge: - PAYMENT-REQUIRED credential: - PAYMENT-SIGNATURE - X-PAYMENT receipt: - PAYMENT-RESPONSE - X-PAYMENT-RESPONSE pricing: amount: '0.001000' currency: USD minimumUsd: '0.001000' mode: fixed quoteBasis: floor protocol: asset: '0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913' network: eip155:8453 payTo: '0x754BD2A6799e766B1E2b9f1Ae7fBdC0CeD96ef13' scheme: exact rail: x402 settlement: onExecutionFailure: released onUnknownFinalization: indeterminate timing: after_execution - headers: challenge: - WWW-Authenticate credential: - Authorization receipt: - Payment-Receipt pricing: amount: '0.000990' currency: USD minimumUsd: null mode: fixed quoteBasis: funded_ceiling protocol: asset: '0x20c000000000000000000000b9537d11c60e8b50' currency: USD intent: charge method: tempo network: tempo payTo: '0x754BD2A6799e766B1E2b9f1Ae7fBdC0CeD96ef13' rail: tempo scheme: exact rail: mpp settlement: onExecutionFailure: charged_undelivered onVerificationError: indeterminate timing: before_execution /v1/run/google.scholar: post: description: 'Search Google Scholar for academic papers with title, authors, citation count, and PDF link. **Price:** \$0.99 per 1,000 requests (flat per request - same cost regardless of results returned). **Routing:** one lane serves this API today, so a failed attempt has nowhere to fail over to. Payment outcome follows the selected rail''s settlement policy. **Catalog:** Google Scholar pricing and uptime - live USD price, lane routing, and measured 30-day uptime. Every Google endpoint.' operationId: google_scholar parameters: - description: Optional. Comma-separated keys (dotted paths like `author.name` descend into nested objects) to keep on each result item. Keys are matched relative to each result item after the `data`/`items` envelope is unwrapped, not against the top-level response envelope, so use `jq` to reshape the whole envelope. Shrinks the response without changing cost. in: query name: fields required: false schema: type: string - description: Optional. Cap the number of result rows returned; a `_truncated` note reports how many were withheld so you can page via the API's own `limit`. Does not change cost. in: query name: max_items required: false schema: minimum: 0 type: integer - description: Optional. Return only a structural outline (top-level keys, item counts, and per-field byte sizes) instead of the full data. Does not change cost. in: query name: summary required: false schema: type: boolean - description: 'Optional. A jq expression applied to the result envelope; its output replaces `output` (multiple outputs collect into an array). Reshape freely, e.g. `jq=.data | {title, description, md: .markdown[:3500]}`. Runs sandboxed with a 250ms / 2MB budget; on failure the full result is returned with a `jqError`. Does not change cost.' in: query name: jq required: false schema: type: string - description: Optional. The most you are willing to pay for this one request, in US dollars (for example 0.05). Any route that would charge more than this is not used, so a request only ever runs on something you can afford. If nothing is available at or below your amount, the request is refused before it runs, nothing is charged, and the message tells you the cheapest price per request so you can raise it. Leave it out to accept the normal price. in: query name: max_cost_usd required: false schema: type: string - description: Optional wallet idempotency key, scoped to this customer for 24 hours. When the gateway honors the key, this synchronous in-process execution can continue after the caller disconnects, bounded by its execution deadline. A completed replayable result charges normally exactly once and can be replayed without another provider run or charge. A pending duplicate returns 409 idempotency_in_progress; reuse with different request semantics returns 409 idempotency_conflict. in: header name: Idempotency-Key required: false schema: maxLength: 255 minLength: 1 type: string requestBody: content: application/json: example: query: attention is all you need schema: additionalProperties: false example: query: attention is all you need properties: preferLatencyUnderMs: description: 'Optional; omit it and routing is unchanged, with the cheapest source serving. Prefer sources whose typical response time (median over the trailing 30 days, as published on this endpoint''s lane health) is under this many milliseconds; among those, the cheapest serves. This can raise your price: when the cheapest source misses the target, a faster and dearer one serves, and you are quoted and charged its price. If no source is that fast the request is still served, by whichever source offers the best speed for its price - it is never refused for being slow. Sources we have not timed are tried last. This is a preference, not a guarantee: the median describes past requests and is not a ceiling on this one, and it excludes any wait this request itself asks for. On a paginated walk it applies to the first page only: later pages stay with the source that page chose, at the price it was quoted.' minimum: 1 type: integer query: description: The Google Scholar search query. type: string required: - query title: Google Scholar input type: object required: true responses: '200': content: application/json: schema: properties: costUsd: description: USD charged on the original run. On a replay this value is echoed for parity; the replay itself is free. type: number hint: description: 'Optional one-line nudge, absent when there is nothing to say. `large_result:` suggests the fields/max_items/summary/jq controls for a big response. `paging_unavailable:` means this result came from a source that cannot return a `nextCursor`, so it may be INCOMPLETE and cannot be continued - re-run with `requireCursor: true` to be served only by a source that can page, which may cost more per request.' type: string items: description: Number of result rows returned. For per-result SKUs the per-item cost is charged against this count; for input-priced SKUs the charge is per submitted input, independent of this count. type: integer jqError: description: Present only when a `jq` expression failed; `output` then carries the full unshaped result and this explains why the reshape did not apply. type: string output: anyOf: - additionalProperties: false properties: data: description: Result payload, or null when nothing was found. oneOf: - type: 'null' - additionalProperties: false properties: query: description: The query that was searched. type: string results: description: Academic paper result records. items: properties: citedBy: description: Number of citations reported by Google Scholar. type: integer id: description: Result identifier. type: string link: description: URL to the paper. Populated whenever the provider has data for the entity. format: uri type: string x-anyapi-must-populate: true pdfUrl: description: URL to an available PDF. format: uri type: string publicationInfo: description: Authors, venue, and publication year. type: string snippet: description: Short paper description snippet. type: string title: description: Paper title. Populated whenever the provider has data for the entity. type: string x-anyapi-must-populate: true year: description: Publication year. type: integer required: - title - link type: object type: array required: - query - results type: object found: description: True when at least one scholar result was returned. type: boolean required: - found - data title: Google Scholar output type: object - type: 'null' description: Normalized output, or null when the replay payload was not retained. provider: description: Always "AnyAPI". type: string replayed: description: True when this response replays the durable result of an earlier run without billing or upstream execution. type: boolean resultId: description: Opaque handle to the full unshaped result, cached ~15 min. Re-shape it for free (fields/max_items/summary/jq) via GET /v1/results/{id}, no re-billing. Absent when the result was too large to cache. type: string required: - output - provider - costUsd - items - replayed type: object description: Normalized result. headers: Idempotency-Replayed: description: Present only when the request carried an Idempotency-Key this gateway honored. "false" means the key was claimed and the run executed freshly; "true" means it replayed a prior identical run and was NOT charged again. Absent means either no key was sent or this gateway does not honor keys, so a retry could be charged twice. schema: enum: - 'false' - 'true' type: string PAYMENT-RESPONSE: description: Inline x402 payment receipt. schema: type: string Payment-Receipt: description: Inline mpp payment receipt. schema: type: string X-PAYMENT-RESPONSE: description: Inline x402 payment receipt. schema: type: string '400': content: application/json: schema: $ref: '#/components/schemas/Error' description: Input did not match the schema, or the upstream rejected the request. '401': content: application/json: schema: $ref: '#/components/schemas/Error' description: Missing or invalid API key. '402': content: application/json: schema: oneOf: - $ref: '#/components/schemas/PaymentRequiredError' - $ref: '#/components/schemas/TrialCapReached' application/problem+json: schema: properties: challengeId: minLength: 1 type: string status: const: 402 type: integer required: - status - challengeId type: object description: Insufficient wallet balance or spend cap, or an inline-payment challenge. Agent trials receive a trial_cap_reached continuation with a live RFC 8628 device authorization. headers: PAYMENT-REQUIRED: description: Inline x402 payment challenge. schema: type: string WWW-Authenticate: description: Inline mpp payment challenge. schema: type: string '404': content: application/json: schema: $ref: '#/components/schemas/Error' description: Unknown SKU, or the upstream had no matching resource. '409': content: application/json: schema: $ref: '#/components/schemas/Error' description: 'Idempotency collision. Codes: idempotency_in_progress means retry after the Retry-After delay; idempotency_conflict means use a new key for different request semantics; idempotency_needs_review means stop retrying and contact AnyAPI support with X-Anyapi-Request-Id.' '422': content: application/json: schema: $ref: '#/components/schemas/Error' description: The request was well formed but this API cannot serve the requested target site; the body's `alternatives` array lists the AnyAPI APIs that can. '429': content: application/json: schema: $ref: '#/components/schemas/Error' description: The upstream was rate limited; inspect `payment` when present before deciding whether to retry. '500': content: application/json: schema: $ref: '#/components/schemas/Error' description: An internal error prevented completion. Inspect payment when present before retrying. '502': content: application/json: schema: $ref: '#/components/schemas/Error' description: Every provider failed with a server/transport error, or inline payment settlement could not be confirmed; inspect `payment` when present before retrying. security: - bearerAuth: [] - apiKeyAuth: [] - {} summary: Google Scholar tags: - Google x-mint: href: /api-reference/google/scholar x-payment-info: price: amount: '0.000990' currency: USD mode: fixed protocols: - x402: asset: '0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913' network: eip155:8453 payTo: '0x754BD2A6799e766B1E2b9f1Ae7fBdC0CeD96ef13' scheme: exact - mpp: asset: '0x20c000000000000000000000b9537d11c60e8b50' currency: USD intent: charge method: tempo network: tempo payTo: '0x754BD2A6799e766B1E2b9f1Ae7fBdC0CeD96ef13' rail: tempo scheme: exact rails: - headers: challenge: - PAYMENT-REQUIRED credential: - PAYMENT-SIGNATURE - X-PAYMENT receipt: - PAYMENT-RESPONSE - X-PAYMENT-RESPONSE pricing: amount: '0.001000' currency: USD minimumUsd: '0.001000' mode: fixed quoteBasis: floor protocol: asset: '0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913' network: eip155:8453 payTo: '0x754BD2A6799e766B1E2b9f1Ae7fBdC0CeD96ef13' scheme: exact rail: x402 settlement: onExecutionFailure: released onUnknownFinalization: indeterminate timing: after_execution - headers: challenge: - WWW-Authenticate credential: - Authorization receipt: - Payment-Receipt pricing: amount: '0.000990' currency: USD minimumUsd: null mode: fixed quoteBasis: funded_ceiling protocol: asset: '0x20c000000000000000000000b9537d11c60e8b50' currency: USD intent: charge method: tempo network: tempo payTo: '0x754BD2A6799e766B1E2b9f1Ae7fBdC0CeD96ef13' rail: tempo scheme: exact rail: mpp settlement: onExecutionFailure: charged_undelivered onVerificationError: indeterminate timing: before_execution /v1/run/google.search: post: description: 'Run a Google web search and get the organic results (title, link, snippet, position) as clean JSON. Returns about 10 results per call - Google stopped honoring bulk result counts in September 2025, so a limit above 10 is accepted but returns no more than a page. Pass the returned nextCursor back as cursor to walk further, or use google.search_100 for up to 100 ranked results in one call, which is cheaper past roughly 20 results. **Price:** \$0.50 per 1,000 requests (flat per request - same cost regardless of results returned). **Routing:** 3 independent lanes are available in execution order. A failed attempt uses the next lane only when the request''s payment authorizes that lane''s price; the normalized schema stays the same. **Catalog:** Google Search pricing and uptime - live USD price, lane routing, and measured 30-day uptime. Every Google endpoint.' operationId: google_search parameters: - description: Optional. Comma-separated keys (dotted paths like `author.name` descend into nested objects) to keep on each result item. Keys are matched relative to each result item after the `data`/`items` envelope is unwrapped, not against the top-level response envelope, so use `jq` to reshape the whole envelope. Shrinks the response without changing cost. in: query name: fields required: false schema: type: string - description: Optional. Cap the number of result rows returned; a `_truncated` note reports how many were withheld so you can page via the API's own `limit`. Does not change cost. in: query name: max_items required: false schema: minimum: 0 type: integer - description: Optional. Return only a structural outline (top-level keys, item counts, and per-field byte sizes) instead of the full data. Does not change cost. in: query name: summary required: false schema: type: boolean - description: 'Optional. A jq expression applied to the result envelope; its output replaces `output` (multiple outputs collect into an array). Reshape freely, e.g. `jq=.data | {title, description, md: .markdown[:3500]}`. Runs sandboxed with a 250ms / 2MB budget; on failure the full result is returned with a `jqError`. Does not change cost.' in: query name: jq required: false schema: type: string - description: Optional. The most you are willing to pay for this one request, in US dollars (for example 0.05). Any route that would charge more than this is not used, so a request only ever runs on something you can afford. If nothing is available at or below your amount, the request is refused before it runs, nothing is charged, and the message tells you the cheapest price per request so you can raise it. Leave it out to accept the normal price. in: query name: max_cost_usd required: false schema: type: string - description: Optional wallet idempotency key, scoped to this customer for 24 hours. When the gateway honors the key, this synchronous in-process execution can continue after the caller disconnects, bounded by its execution deadline. A completed replayable result charges normally exactly once and can be replayed without another provider run or charge. A pending duplicate returns 409 idempotency_in_progress; reuse with different request semantics returns 409 idempotency_conflict. in: header name: Idempotency-Key required: false schema: maxLength: 255 minLength: 1 type: string requestBody: content: application/json: example: gl: us hl: en limit: 10 query: best coffee maker schema: additionalProperties: false example: gl: us hl: en limit: 10 query: best coffee maker properties: autocorrect: description: Toggle Google spelling autocorrect (default true). Set false to search the exact query without correction. type: boolean cursor: description: Continuation token from a previous response's nextCursor. Pass it back to fetch the next page of results. minLength: 1 type: string gl: default: us description: Two-letter country code for result localization (e.g. us, gb, de). type: string hl: default: en description: Two-letter interface and results language code (e.g. en, es, de). type: string limit: default: 10 description: Maximum number of organic results to return in this response. Google stopped honoring bulk result counts in September 2025, so one page is about 10 results and a limit above 10 is accepted but will not return more than that. To go deeper, either page through with cursor (about 10 results per call, each billed as a request) or use google.search_100, which returns up to 100 ranked results in a single call for one flat charge and is cheaper past roughly 20 results. Price is flat per request. maximum: 100 minimum: 1 type: integer location: description: Fine-grained location for result localization, given as a canonical Google location string (e.g. 'New York, United States', 'London, United Kingdom'). More specific than the country-level gl. type: string preferLatencyUnderMs: description: 'Optional; omit it and routing is unchanged, with the cheapest source serving. Prefer sources whose typical response time (median over the trailing 30 days, as published on this endpoint''s lane health) is under this many milliseconds; among those, the cheapest serves. This can raise your price: when the cheapest source misses the target, a faster and dearer one serves, and you are quoted and charged its price. If no source is that fast the request is still served, by whichever source offers the best speed for its price - it is never refused for being slow. Sources we have not timed are tried last. This is a preference, not a guarantee: the median describes past requests and is not a ceiling on this one, and it excludes any wait this request itself asks for. On a paginated walk it applies to the first page only: later pages stay with the source that page chose, at the price it was quoted.' minimum: 1 type: integer query: description: The Google search query. type: string requireCursor: description: Set true if you intend to page through results. The cheapest source for this search cannot return a nextCursor, so by default a single call may come back with no way to continue; setting this routes to a source that can page, at a higher price per request. type: boolean timeframe: description: 'Restrict results to a recent time window: 1h, 1d, 7d, 1y, or all. Default all (no time restriction).' type: string required: - query title: Google search input type: object required: true responses: '200': content: application/json: schema: properties: costUsd: description: USD charged on the original run. On a replay this value is echoed for parity; the replay itself is free. type: number hint: description: 'Optional one-line nudge, absent when there is nothing to say. `large_result:` suggests the fields/max_items/summary/jq controls for a big response. `paging_unavailable:` means this result came from a source that cannot return a `nextCursor`, so it may be INCOMPLETE and cannot be continued - re-run with `requireCursor: true` to be served only by a source that can page, which may cost more per request.' type: string items: description: Number of result rows returned. For per-result SKUs the per-item cost is charged against this count; for input-priced SKUs the charge is per submitted input, independent of this count. type: integer jqError: description: Present only when a `jq` expression failed; `output` then carries the full unshaped result and this explains why the reshape did not apply. type: string output: anyOf: - additionalProperties: false properties: data: oneOf: - type: 'null' - additionalProperties: false properties: nextCursor: description: Opaque cursor for the next page of results, or null when this lane has no more. Pass it back as cursor to continue. type: - string - 'null' query: type: string results: description: Populated whenever the provider has data for the entity. items: properties: link: description: Populated whenever the provider has data for the entity. type: string x-anyapi-must-populate: true position: type: integer snippet: description: Populated whenever the provider has data for the entity. type: string x-anyapi-must-populate: true title: description: Populated whenever the provider has data for the entity. type: string x-anyapi-must-populate: true required: - title - link - snippet - position type: object type: array x-anyapi-must-populate: true required: - query - results type: object found: type: boolean required: - found - data title: Google search output type: object - type: 'null' description: Normalized output, or null when the replay payload was not retained. provider: description: Always "AnyAPI". type: string replayed: description: True when this response replays the durable result of an earlier run without billing or upstream execution. type: boolean resultId: description: Opaque handle to the full unshaped result, cached ~15 min. Re-shape it for free (fields/max_items/summary/jq) via GET /v1/results/{id}, no re-billing. Absent when the result was too large to cache. type: string required: - output - provider - costUsd - items - replayed type: object description: Normalized result. headers: Idempotency-Replayed: description: Present only when the request carried an Idempotency-Key this gateway honored. "false" means the key was claimed and the run executed freshly; "true" means it replayed a prior identical run and was NOT charged again. Absent means either no key was sent or this gateway does not honor keys, so a retry could be charged twice. schema: enum: - 'false' - 'true' type: string PAYMENT-RESPONSE: description: Inline x402 payment receipt. schema: type: string Payment-Receipt: description: Inline mpp payment receipt. schema: type: string X-PAYMENT-RESPONSE: description: Inline x402 payment receipt. schema: type: string '400': content: application/json: schema: $ref: '#/components/schemas/Error' description: Input did not match the schema, or the upstream rejected the request. '401': content: application/json: schema: $ref: '#/components/schemas/Error' description: Missing or invalid API key. '402': content: application/json: schema: oneOf: - $ref: '#/components/schemas/PaymentRequiredError' - $ref: '#/components/schemas/TrialCapReached' application/problem+json: schema: properties: challengeId: minLength: 1 type: string status: const: 402 type: integer required: - status - challengeId type: object description: Insufficient wallet balance or spend cap, or an inline-payment challenge. Agent trials receive a trial_cap_reached continuation with a live RFC 8628 device authorization. headers: PAYMENT-REQUIRED: description: Inline x402 payment challenge. schema: type: string WWW-Authenticate: description: Inline mpp payment challenge. schema: type: string '404': content: application/json: schema: $ref: '#/components/schemas/Error' description: Unknown SKU, or the upstream had no matching resource. '409': content: application/json: schema: $ref: '#/components/schemas/Error' description: 'Idempotency collision. Codes: idempotency_in_progress means retry after the Retry-After delay; idempotency_conflict means use a new key for different request semantics; idempotency_needs_review means stop retrying and contact AnyAPI support with X-Anyapi-Request-Id.' '422': content: application/json: schema: $ref: '#/components/schemas/Error' description: The request was well formed but this API cannot serve the requested target site; the body's `alternatives` array lists the AnyAPI APIs that can. '429': content: application/json: schema: $ref: '#/components/schemas/Error' description: The upstream was rate limited; inspect `payment` when present before deciding whether to retry. '500': content: application/json: schema: $ref: '#/components/schemas/Error' description: An internal error prevented completion. Inspect payment when present before retrying. '502': content: application/json: schema: $ref: '#/components/schemas/Error' description: Every provider failed with a server/transport error, or inline payment settlement could not be confirmed; inspect `payment` when present before retrying. security: - bearerAuth: [] - apiKeyAuth: [] - {} summary: Google Search tags: - Google x-mint: href: /api-reference/google/search x-payment-info: price: currency: USD max: '0.001260' min: '0.000500' mode: dynamic protocols: - x402: asset: '0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913' network: eip155:8453 payTo: '0x754BD2A6799e766B1E2b9f1Ae7fBdC0CeD96ef13' scheme: exact - mpp: asset: '0x20c000000000000000000000b9537d11c60e8b50' currency: USD intent: charge method: tempo network: tempo payTo: '0x754BD2A6799e766B1E2b9f1Ae7fBdC0CeD96ef13' rail: tempo scheme: exact rails: - headers: challenge: - PAYMENT-REQUIRED credential: - PAYMENT-SIGNATURE - X-PAYMENT receipt: - PAYMENT-RESPONSE - X-PAYMENT-RESPONSE pricing: currency: USD max: '0.001260' min: '0.001000' minimumUsd: '0.001000' mode: dynamic quoteBasis: floor protocol: asset: '0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913' network: eip155:8453 payTo: '0x754BD2A6799e766B1E2b9f1Ae7fBdC0CeD96ef13' scheme: exact rail: x402 settlement: onExecutionFailure: released onUnknownFinalization: indeterminate timing: after_execution - headers: challenge: - WWW-Authenticate credential: - Authorization receipt: - Payment-Receipt pricing: currency: USD max: '0.001260' min: '0.000500' minimumUsd: null mode: dynamic quoteBasis: funded_ceiling protocol: asset: '0x20c000000000000000000000b9537d11c60e8b50' currency: USD intent: charge method: tempo network: tempo payTo: '0x754BD2A6799e766B1E2b9f1Ae7fBdC0CeD96ef13' rail: tempo scheme: exact rail: mpp settlement: onExecutionFailure: charged_undelivered onVerificationError: indeterminate timing: before_execution /v1/run/google.search_100: post: description: 'Run a Google web search and get up to 100 ranked organic results in one call, with true absolute positions rather than per-page numbering. One flat charge whatever the depth, which makes it cheaper than paging google.search past roughly 20 results. It reads ten pages of Google to build the list, so a call takes around three to four minutes - use google.search when you want the first page back in a second. **Price:** \$1.80 per 1,000 requests (flat per request - same cost regardless of results returned). **Routing:** one lane serves this API today, so a failed attempt has nowhere to fail over to. Payment outcome follows the selected rail''s settlement policy. **Catalog:** Google Search Top 100 pricing and uptime - live USD price, lane routing, and measured 30-day uptime. Every Google endpoint.' operationId: google_search_100 parameters: - description: Optional. Comma-separated keys (dotted paths like `author.name` descend into nested objects) to keep on each result item. Keys are matched relative to each result item after the `data`/`items` envelope is unwrapped, not against the top-level response envelope, so use `jq` to reshape the whole envelope. Shrinks the response without changing cost. in: query name: fields required: false schema: type: string - description: Optional. Cap the number of result rows returned; a `_truncated` note reports how many were withheld so you can page via the API's own `limit`. Does not change cost. in: query name: max_items required: false schema: minimum: 0 type: integer - description: Optional. Return only a structural outline (top-level keys, item counts, and per-field byte sizes) instead of the full data. Does not change cost. in: query name: summary required: false schema: type: boolean - description: 'Optional. A jq expression applied to the result envelope; its output replaces `output` (multiple outputs collect into an array). Reshape freely, e.g. `jq=.data | {title, description, md: .markdown[:3500]}`. Runs sandboxed with a 250ms / 2MB budget; on failure the full result is returned with a `jqError`. Does not change cost.' in: query name: jq required: false schema: type: string - description: Optional. The most you are willing to pay for this one request, in US dollars (for example 0.05). Any route that would charge more than this is not used, so a request only ever runs on something you can afford. If nothing is available at or below your amount, the request is refused before it runs, nothing is charged, and the message tells you the cheapest price per request so you can raise it. Leave it out to accept the normal price. in: query name: max_cost_usd required: false schema: type: string - description: Optional wallet idempotency key, scoped to this customer for 24 hours. When the gateway honors the key, this synchronous in-process execution can continue after the caller disconnects, bounded by its execution deadline. A completed replayable result charges normally exactly once and can be replayed without another provider run or charge. A pending duplicate returns 409 idempotency_in_progress; reuse with different request semantics returns 409 idempotency_conflict. in: header name: Idempotency-Key required: false schema: maxLength: 255 minLength: 1 type: string requestBody: content: application/json: example: gl: us hl: en query: best crm software schema: additionalProperties: false example: gl: us hl: en query: best crm software properties: autocorrect: description: Toggle Google spelling autocorrect (default true). Set false to search the exact query without correction. type: boolean gl: default: us description: Two-letter country code for result localization (e.g. us, gb, de). type: string hl: default: en description: Two-letter interface and results language code (e.g. en, es, de). type: string preferLatencyUnderMs: description: 'Optional; omit it and routing is unchanged, with the cheapest source serving. Prefer sources whose typical response time (median over the trailing 30 days, as published on this endpoint''s lane health) is under this many milliseconds; among those, the cheapest serves. This can raise your price: when the cheapest source misses the target, a faster and dearer one serves, and you are quoted and charged its price. If no source is that fast the request is still served, by whichever source offers the best speed for its price - it is never refused for being slow. Sources we have not timed are tried last. This is a preference, not a guarantee: the median describes past requests and is not a ceiling on this one, and it excludes any wait this request itself asks for. On a paginated walk it applies to the first page only: later pages stay with the source that page chose, at the price it was quoted.' minimum: 1 type: integer query: description: The Google search query. type: string timeframe: description: 'Restrict results to a recent time window: 1h, 1d, 7d, 1y, or all. Default all (no time restriction).' type: string required: - query title: Google search top 100 input type: object required: true responses: '200': content: application/json: schema: properties: costUsd: description: USD charged on the original run. On a replay this value is echoed for parity; the replay itself is free. type: number hint: description: 'Optional one-line nudge, absent when there is nothing to say. `large_result:` suggests the fields/max_items/summary/jq controls for a big response. `paging_unavailable:` means this result came from a source that cannot return a `nextCursor`, so it may be INCOMPLETE and cannot be continued - re-run with `requireCursor: true` to be served only by a source that can page, which may cost more per request.' type: string items: description: Number of result rows returned. For per-result SKUs the per-item cost is charged against this count; for input-priced SKUs the charge is per submitted input, independent of this count. type: integer jqError: description: Present only when a `jq` expression failed; `output` then carries the full unshaped result and this explains why the reshape did not apply. type: string output: anyOf: - additionalProperties: false properties: data: oneOf: - type: 'null' - additionalProperties: false properties: aiOverview: description: Google's AI Overview text for this query, when Google showed one. Absent when it did not. type: string query: description: The search query these results answer. type: string results: description: Organic results in Google's own order, position 1 first. Populated whenever the provider has data for the entity. items: properties: link: description: The destination URL. Populated whenever the provider has data for the entity. type: string x-anyapi-must-populate: true position: description: Absolute rank across the whole result set, counting from 1 - not restarted per page. Populated whenever the provider has data for the entity. type: integer x-anyapi-must-populate: true snippet: description: Google's summary text for the result. Populated whenever the provider has data for the entity. type: string x-anyapi-must-populate: true title: description: The result's headline as Google renders it. Populated whenever the provider has data for the entity. type: string x-anyapi-must-populate: true required: - title - link - snippet - position type: object type: array x-anyapi-must-populate: true required: - query - results type: object found: type: boolean required: - found - data title: Google search top 100 output type: object - type: 'null' description: Normalized output, or null when the replay payload was not retained. provider: description: Always "AnyAPI". type: string replayed: description: True when this response replays the durable result of an earlier run without billing or upstream execution. type: boolean resultId: description: Opaque handle to the full unshaped result, cached ~15 min. Re-shape it for free (fields/max_items/summary/jq) via GET /v1/results/{id}, no re-billing. Absent when the result was too large to cache. type: string required: - output - provider - costUsd - items - replayed type: object description: Normalized result. headers: Idempotency-Replayed: description: Present only when the request carried an Idempotency-Key this gateway honored. "false" means the key was claimed and the run executed freshly; "true" means it replayed a prior identical run and was NOT charged again. Absent means either no key was sent or this gateway does not honor keys, so a retry could be charged twice. schema: enum: - 'false' - 'true' type: string PAYMENT-RESPONSE: description: Inline x402 payment receipt. schema: type: string Payment-Receipt: description: Inline mpp payment receipt. schema: type: string X-PAYMENT-RESPONSE: description: Inline x402 payment receipt. schema: type: string '400': content: application/json: schema: $ref: '#/components/schemas/Error' description: Input did not match the schema, or the upstream rejected the request. '401': content: application/json: schema: $ref: '#/components/schemas/Error' description: Missing or invalid API key. '402': content: application/json: schema: oneOf: - $ref: '#/components/schemas/PaymentRequiredError' - $ref: '#/components/schemas/TrialCapReached' application/problem+json: schema: properties: challengeId: minLength: 1 type: string status: const: 402 type: integer required: - status - challengeId type: object description: Insufficient wallet balance or spend cap, or an inline-payment challenge. Agent trials receive a trial_cap_reached continuation with a live RFC 8628 device authorization. headers: PAYMENT-REQUIRED: description: Inline x402 payment challenge. schema: type: string WWW-Authenticate: description: Inline mpp payment challenge. schema: type: string '404': content: application/json: schema: $ref: '#/components/schemas/Error' description: Unknown SKU, or the upstream had no matching resource. '409': content: application/json: schema: $ref: '#/components/schemas/Error' description: 'Idempotency collision. Codes: idempotency_in_progress means retry after the Retry-After delay; idempotency_conflict means use a new key for different request semantics; idempotency_needs_review means stop retrying and contact AnyAPI support with X-Anyapi-Request-Id.' '422': content: application/json: schema: $ref: '#/components/schemas/Error' description: The request was well formed but this API cannot serve the requested target site; the body's `alternatives` array lists the AnyAPI APIs that can. '429': content: application/json: schema: $ref: '#/components/schemas/Error' description: The upstream was rate limited; inspect `payment` when present before deciding whether to retry. '500': content: application/json: schema: $ref: '#/components/schemas/Error' description: An internal error prevented completion. Inspect payment when present before retrying. '502': content: application/json: schema: $ref: '#/components/schemas/Error' description: Every provider failed with a server/transport error, or inline payment settlement could not be confirmed; inspect `payment` when present before retrying. security: - bearerAuth: [] - apiKeyAuth: [] - {} summary: Google Search Top 100 tags: - Google x-mint: href: /api-reference/google/search-100 x-payment-info: price: amount: '0.001800' currency: USD mode: fixed protocols: - x402: asset: '0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913' network: eip155:8453 payTo: '0x754BD2A6799e766B1E2b9f1Ae7fBdC0CeD96ef13' scheme: exact - mpp: asset: '0x20c000000000000000000000b9537d11c60e8b50' currency: USD intent: charge method: tempo network: tempo payTo: '0x754BD2A6799e766B1E2b9f1Ae7fBdC0CeD96ef13' rail: tempo scheme: exact rails: - headers: challenge: - PAYMENT-REQUIRED credential: - PAYMENT-SIGNATURE - X-PAYMENT receipt: - PAYMENT-RESPONSE - X-PAYMENT-RESPONSE pricing: amount: '0.001800' currency: USD minimumUsd: '0.001000' mode: fixed quoteBasis: floor protocol: asset: '0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913' network: eip155:8453 payTo: '0x754BD2A6799e766B1E2b9f1Ae7fBdC0CeD96ef13' scheme: exact rail: x402 settlement: onExecutionFailure: released onUnknownFinalization: indeterminate timing: after_execution - headers: challenge: - WWW-Authenticate credential: - Authorization receipt: - Payment-Receipt pricing: amount: '0.001800' currency: USD minimumUsd: null mode: fixed quoteBasis: funded_ceiling protocol: asset: '0x20c000000000000000000000b9537d11c60e8b50' currency: USD intent: charge method: tempo network: tempo payTo: '0x754BD2A6799e766B1E2b9f1Ae7fBdC0CeD96ef13' rail: tempo scheme: exact rail: mpp settlement: onExecutionFailure: charged_undelivered onVerificationError: indeterminate timing: before_execution /v1/run/google.videos: post: description: 'Search Google for video results (YouTube and others) with title, link, thumbnail, and source. **Price:** \$0.99 per 1,000 requests (flat per request - same cost regardless of results returned). **Routing:** one lane serves this API today, so a failed attempt has nowhere to fail over to. Payment outcome follows the selected rail''s settlement policy. **Catalog:** Google Videos pricing and uptime - live USD price, lane routing, and measured 30-day uptime. Every Google endpoint.' operationId: google_videos parameters: - description: Optional. Comma-separated keys (dotted paths like `author.name` descend into nested objects) to keep on each result item. Keys are matched relative to each result item after the `data`/`items` envelope is unwrapped, not against the top-level response envelope, so use `jq` to reshape the whole envelope. Shrinks the response without changing cost. in: query name: fields required: false schema: type: string - description: Optional. Cap the number of result rows returned; a `_truncated` note reports how many were withheld so you can page via the API's own `limit`. Does not change cost. in: query name: max_items required: false schema: minimum: 0 type: integer - description: Optional. Return only a structural outline (top-level keys, item counts, and per-field byte sizes) instead of the full data. Does not change cost. in: query name: summary required: false schema: type: boolean - description: 'Optional. A jq expression applied to the result envelope; its output replaces `output` (multiple outputs collect into an array). Reshape freely, e.g. `jq=.data | {title, description, md: .markdown[:3500]}`. Runs sandboxed with a 250ms / 2MB budget; on failure the full result is returned with a `jqError`. Does not change cost.' in: query name: jq required: false schema: type: string - description: Optional. The most you are willing to pay for this one request, in US dollars (for example 0.05). Any route that would charge more than this is not used, so a request only ever runs on something you can afford. If nothing is available at or below your amount, the request is refused before it runs, nothing is charged, and the message tells you the cheapest price per request so you can raise it. Leave it out to accept the normal price. in: query name: max_cost_usd required: false schema: type: string - description: Optional wallet idempotency key, scoped to this customer for 24 hours. When the gateway honors the key, this synchronous in-process execution can continue after the caller disconnects, bounded by its execution deadline. A completed replayable result charges normally exactly once and can be replayed without another provider run or charge. A pending duplicate returns 409 idempotency_in_progress; reuse with different request semantics returns 409 idempotency_conflict. in: header name: Idempotency-Key required: false schema: maxLength: 255 minLength: 1 type: string requestBody: content: application/json: example: gl: us hl: en query: lofi hip hop schema: additionalProperties: false example: gl: us hl: en query: lofi hip hop properties: autocorrect: description: Toggle Google spelling autocorrect (default true). Set false to search the exact query without correction. type: boolean gl: default: us description: Two-letter country code for result localization (e.g. us, gb, de). type: string hl: default: en description: Two-letter interface and results language code (e.g. en, es, de). type: string location: description: Fine-grained location for result localization, given as a canonical Google location string (e.g. 'New York, United States', 'London, United Kingdom'). More specific than the country-level gl. type: string preferLatencyUnderMs: description: 'Optional; omit it and routing is unchanged, with the cheapest source serving. Prefer sources whose typical response time (median over the trailing 30 days, as published on this endpoint''s lane health) is under this many milliseconds; among those, the cheapest serves. This can raise your price: when the cheapest source misses the target, a faster and dearer one serves, and you are quoted and charged its price. If no source is that fast the request is still served, by whichever source offers the best speed for its price - it is never refused for being slow. Sources we have not timed are tried last. This is a preference, not a guarantee: the median describes past requests and is not a ceiling on this one, and it excludes any wait this request itself asks for. On a paginated walk it applies to the first page only: later pages stay with the source that page chose, at the price it was quoted.' minimum: 1 type: integer query: description: The video search query. type: string timeframe: description: 'Restrict results to a recent time window: 1h, 1d, 7d, 1y, or all. Default all (no time restriction).' type: string required: - query title: Google Videos input type: object required: true responses: '200': content: application/json: schema: properties: costUsd: description: USD charged on the original run. On a replay this value is echoed for parity; the replay itself is free. type: number hint: description: 'Optional one-line nudge, absent when there is nothing to say. `large_result:` suggests the fields/max_items/summary/jq controls for a big response. `paging_unavailable:` means this result came from a source that cannot return a `nextCursor`, so it may be INCOMPLETE and cannot be continued - re-run with `requireCursor: true` to be served only by a source that can page, which may cost more per request.' type: string items: description: Number of result rows returned. For per-result SKUs the per-item cost is charged against this count; for input-priced SKUs the charge is per submitted input, independent of this count. type: integer jqError: description: Present only when a `jq` expression failed; `output` then carries the full unshaped result and this explains why the reshape did not apply. type: string output: anyOf: - additionalProperties: false properties: data: description: Result payload, or null when nothing was found. oneOf: - type: 'null' - additionalProperties: false properties: query: description: The query that was searched. type: string results: description: Video result records. Populated whenever the provider has data for the entity. items: properties: image: description: Thumbnail image URL for the video. format: uri type: string link: description: URL to the video. Populated whenever the provider has data for the entity. format: uri type: string x-anyapi-must-populate: true position: description: 1-based rank in the result list. type: integer snippet: description: Short description snippet. type: string source: description: Host platform (e.g. YouTube, Vimeo). type: string title: description: Video title. Populated whenever the provider has data for the entity. type: string x-anyapi-must-populate: true required: - title - link type: object type: array x-anyapi-must-populate: true required: - query - results type: object found: description: True when at least one video result was returned. type: boolean required: - found - data title: Google Videos output type: object - type: 'null' description: Normalized output, or null when the replay payload was not retained. provider: description: Always "AnyAPI". type: string replayed: description: True when this response replays the durable result of an earlier run without billing or upstream execution. type: boolean resultId: description: Opaque handle to the full unshaped result, cached ~15 min. Re-shape it for free (fields/max_items/summary/jq) via GET /v1/results/{id}, no re-billing. Absent when the result was too large to cache. type: string required: - output - provider - costUsd - items - replayed type: object description: Normalized result. headers: Idempotency-Replayed: description: Present only when the request carried an Idempotency-Key this gateway honored. "false" means the key was claimed and the run executed freshly; "true" means it replayed a prior identical run and was NOT charged again. Absent means either no key was sent or this gateway does not honor keys, so a retry could be charged twice. schema: enum: - 'false' - 'true' type: string PAYMENT-RESPONSE: description: Inline x402 payment receipt. schema: type: string Payment-Receipt: description: Inline mpp payment receipt. schema: type: string X-PAYMENT-RESPONSE: description: Inline x402 payment receipt. schema: type: string '400': content: application/json: schema: $ref: '#/components/schemas/Error' description: Input did not match the schema, or the upstream rejected the request. '401': content: application/json: schema: $ref: '#/components/schemas/Error' description: Missing or invalid API key. '402': content: application/json: schema: oneOf: - $ref: '#/components/schemas/PaymentRequiredError' - $ref: '#/components/schemas/TrialCapReached' application/problem+json: schema: properties: challengeId: minLength: 1 type: string status: const: 402 type: integer required: - status - challengeId type: object description: Insufficient wallet balance or spend cap, or an inline-payment challenge. Agent trials receive a trial_cap_reached continuation with a live RFC 8628 device authorization. headers: PAYMENT-REQUIRED: description: Inline x402 payment challenge. schema: type: string WWW-Authenticate: description: Inline mpp payment challenge. schema: type: string '404': content: application/json: schema: $ref: '#/components/schemas/Error' description: Unknown SKU, or the upstream had no matching resource. '409': content: application/json: schema: $ref: '#/components/schemas/Error' description: 'Idempotency collision. Codes: idempotency_in_progress means retry after the Retry-After delay; idempotency_conflict means use a new key for different request semantics; idempotency_needs_review means stop retrying and contact AnyAPI support with X-Anyapi-Request-Id.' '422': content: application/json: schema: $ref: '#/components/schemas/Error' description: The request was well formed but this API cannot serve the requested target site; the body's `alternatives` array lists the AnyAPI APIs that can. '429': content: application/json: schema: $ref: '#/components/schemas/Error' description: The upstream was rate limited; inspect `payment` when present before deciding whether to retry. '500': content: application/json: schema: $ref: '#/components/schemas/Error' description: An internal error prevented completion. Inspect payment when present before retrying. '502': content: application/json: schema: $ref: '#/components/schemas/Error' description: Every provider failed with a server/transport error, or inline payment settlement could not be confirmed; inspect `payment` when present before retrying. security: - bearerAuth: [] - apiKeyAuth: [] - {} summary: Google Videos tags: - Google x-mint: href: /api-reference/google/videos x-payment-info: price: amount: '0.000990' currency: USD mode: fixed protocols: - x402: asset: '0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913' network: eip155:8453 payTo: '0x754BD2A6799e766B1E2b9f1Ae7fBdC0CeD96ef13' scheme: exact - mpp: asset: '0x20c000000000000000000000b9537d11c60e8b50' currency: USD intent: charge method: tempo network: tempo payTo: '0x754BD2A6799e766B1E2b9f1Ae7fBdC0CeD96ef13' rail: tempo scheme: exact rails: - headers: challenge: - PAYMENT-REQUIRED credential: - PAYMENT-SIGNATURE - X-PAYMENT receipt: - PAYMENT-RESPONSE - X-PAYMENT-RESPONSE pricing: amount: '0.001000' currency: USD minimumUsd: '0.001000' mode: fixed quoteBasis: floor protocol: asset: '0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913' network: eip155:8453 payTo: '0x754BD2A6799e766B1E2b9f1Ae7fBdC0CeD96ef13' scheme: exact rail: x402 settlement: onExecutionFailure: released onUnknownFinalization: indeterminate timing: after_execution - headers: challenge: - WWW-Authenticate credential: - Authorization receipt: - Payment-Receipt pricing: amount: '0.000990' currency: USD minimumUsd: null mode: fixed quoteBasis: funded_ceiling protocol: asset: '0x20c000000000000000000000b9537d11c60e8b50' currency: USD intent: charge method: tempo network: tempo payTo: '0x754BD2A6799e766B1E2b9f1Ae7fBdC0CeD96ef13' rail: tempo scheme: exact rail: mpp settlement: onExecutionFailure: charged_undelivered onVerificationError: indeterminate timing: before_execution components: schemas: AgentUpgrade: properties: authorizationServer: format: uri type: string clientId: type: string device: $ref: '#/components/schemas/DeviceUpgrade' scope: type: string required: - authorizationServer - scope type: object Payment: properties: costUsd: description: Known USD payment amount for this request. minimum: 0 type: number rail: minLength: 1 type: string settlementState: enum: - charged_undelivered - indeterminate type: string required: - rail - settlementState - costUsd type: object Error: properties: code: description: Stable machine-readable error code when the endpoint defines one. type: string error: description: Customer-safe error message. type: string payment: $ref: '#/components/schemas/Payment' requestId: description: This run's AnyAPI request id, the same value as the X-Anyapi-Request-Id response header. Quote it to support. Absent on endpoints that do not execute a run. format: uuid type: string required: - error type: object PaymentRequiredError: allOf: - $ref: '#/components/schemas/Error' - properties: error: not: const: trial_cap_reached type: object DeviceUpgrade: properties: deviceCode: type: string expiresIn: minimum: 0 type: integer grantType: type: string interval: minimum: 1 type: integer tokenEndpoint: format: uri type: string userCode: type: string verificationUri: format: uri type: string verificationUriComplete: format: uri type: string required: - verificationUriComplete - verificationUri - userCode - deviceCode - tokenEndpoint - grantType - interval - expiresIn type: object TrialCapReached: properties: error: const: trial_cap_reached type: string message: description: Human-relayable continuation instructions, including an executable token poll. type: string upgrade: $ref: '#/components/schemas/AgentUpgrade' required: - error - message - upgrade type: object securitySchemes: apiKeyAuth: description: Your AnyAPI key. in: header name: X-API-Key type: apiKey x-default: YOUR_ANYAPI_KEY bearerAuth: description: Your AnyAPI key as a Bearer token. scheme: bearer type: http x-default: YOUR_ANYAPI_KEY