> ## Documentation Index
> Fetch the complete documentation index at: https://composio-preview.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# CapSolver

> CapSolver provides APIs for CAPTCHA recognition and token-solving workflows, including task creation, polling, account balance, and solution feedback.

![CapSolver logo](https://mintcdn.com/composio-preview/PudKL42NVsy5O-ND/images/toolkits/capsolver.svg?fit=max\&auto=format\&n=PudKL42NVsy5O-ND\&q=85\&s=d436352b34c1795d8946d6b0d9900019)

* **Category:** developer tools
* **Auth:** API\_KEY
* **Composio-managed OAuth available?** N/A
* **Tools:** 16
* **Triggers:** 0
* **Slug:** `CAPSOLVER`
* **Version:** 00000000\_00

## Tools

> [accordion] Classify AWS WAF Image
>
> **Slug:** `CAPSOLVER_CLASSIFY_AWS_WAF`
>
> Classify one AWS WAF challenge image or a nine-image grid and return the immediate geometric or object result. This is a paid CapSolver operation.
>
> #### Input Parameters
>
> **app\_id** `string`
>
> Optional CapSolver application identifier for usage attribution.
>
> **question** `string ("aws:toycarcity:carcity" | "aws:grid:bag" | "aws:grid:bed" | "aws:grid:belts" | "aws:grid:binocular" | "aws:grid:bucket" | "aws:grid:chair" | "aws:grid:clock" | "aws:grid:cooking pot" | "aws:grid:curtain" | "aws:grid:fork" | "aws:grid:hat" | "aws:grid:scissors" | "aws:grid:seat" | "aws:grid:spoon" | "aws:grid:suitcase" | "aws:grid:umbrella")` *required*
>
> Provider-documented AWS WAF challenge question identifier.
>
> **website\_url** `string`
>
> Optional URL of the page containing the challenge.
>
> **images\_base64** `array` *required*
>
> Base64 challenge images. Supply exactly nine images for an aws:grid question and exactly one for aws:toycarcity:carcity.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Classify reCAPTCHA V2 Image
>
> **Slug:** `CAPSOLVER_CLASSIFY_RECAPTCHA_V2`
>
> Classify a reCAPTCHA v2 challenge image for a documented object category and return the immediate selection result. This is a paid CapSolver operation.
>
> #### Input Parameters
>
> **app\_id** `string`
>
> Optional CapSolver application identifier for usage attribution.
>
> **question** `string ("taxis" | "bus" | "school_bus" | "motorcycles" | "tractors" | "chimneys" | "crosswalks" | "traffic_lights" | "bicycles" | "parking_meters" | "cars" | "bridges" | "boats" | "palm_trees" | "mountains_or_hills" | "fire_hydrant" | "stairs")` *required*
>
> Object category named by the challenge prompt. Choose from taxis, bus (generic buses), school\_bus (school buses), motorcycles, tractors, chimneys, crosswalks, traffic\_lights, bicycles, parking\_meters, cars, bridges, boats, palm\_trees, mountains\_or\_hills, fire\_hydrant, or stairs.
>
> **website\_key** `string`
>
> Optional reCAPTCHA site key associated with the challenge.
>
> **website\_url** `string`
>
> Optional URL of the page containing the challenge.
>
> **image\_base64** `string` *required*
>
> Base64-encoded reCAPTCHA challenge image.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Get Balance
>
> **Slug:** `CAPSOLVER_GET_BALANCE`
>
> Return the connected CapSolver account balance and non-secret package metadata. Package access tokens are never returned.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Get Task Result
>
> **Slug:** `CAPSOLVER_GET_TASK_RESULT`
>
> Get one asynchronous CapSolver task result, or optionally poll it every three seconds for a bounded time. This action never creates or retries a task.
>
> #### Input Parameters
>
> **task\_id** `string` *required*
>
> Task identifier returned by a CapSolver task-creation operation.
>
> **wait\_until\_ready** `boolean`
>
> Poll every three seconds until the task is ready or the timeout expires. False performs exactly one result request.
>
> **wait\_timeout\_seconds** `integer`
>
> Window for scheduling polls (3-300 seconds). A final scheduled request may use its network timeout; then the latest state and task ID are returned.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Run Vision Engine
>
> **Slug:** `CAPSOLVER_RUN_VISION_ENGINE`
>
> Run one documented CapSolver vision module and return its immediate points, distance, angle, rectangles, or text result. This is a paid operation.
>
> #### Input Parameters
>
> **app\_id** `string`
>
> Optional CapSolver application identifier for usage attribution.
>
> **module** `string ("botdeflector" | "slider_1" | "rotate_1" | "rotate_2" | "shein" | "ocr_gif")` *required*
>
> Provider module and contract: botdeflector requires image\_base64 and background\_image\_base64 and returns points as x/y pairs; slider\_1 requires both images and returns distance; rotate\_1 requires both images and returns an angle; rotate\_2 uses image\_base64 and returns an angle; shein requires image\_base64 and question and returns ordered rectangles; ocr\_gif uses a base64 GIF and returns recognized text. The selected value is sent to CapSolver unchanged.
>
> **question** `string`
>
> Challenge question. Required only for the shein module.
>
> **website\_url** `string`
>
> Optional URL of the page containing the visual challenge.
>
> **image\_base64** `string` *required*
>
> Raw base64 foreground or challenge image without a data-URI prefix or newlines.
>
> **background\_image\_base64** `string`
>
> Raw base64 background image. Required for botdeflector, slider\_1, and rotate\_1; optional for the other modules.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Solve AWS WAF
>
> **Slug:** `CAPSOLVER_SOLVE_AWS_WAF`
>
> Create one AWS WAF token task with or without a caller-supplied proxy and optionally poll every three seconds for a bounded time. Returns the task ID when processing continues. This is a paid CapSolver operation.
>
> #### Input Parameters
>
> **proxy** `string`
>
> Caller-controlled proxy in a CapSolver-supported format. Required for variant=proxy and never returned.
>
> **app\_id** `string`
>
> Optional CapSolver application identifier for usage attribution.
>
> **aws\_iv** `string`
>
> Sensitive AWS WAF initialization vector from the challenge data.
>
> **aws\_key** `string`
>
> AWS WAF key extracted from the challenge data, when available.
>
> **variant** `string ("proxyless" | "proxy")`
>
> Whether CapSolver connects directly or through the supplied proxy.
>
> **aws\_api\_js** `string`
>
> Sensitive AWS API JavaScript URL documented by CapSolver.
>
> **aws\_api\_key** `string`
>
> Sensitive challenge-specific AWS API key; never returned.
>
> **aws\_context** `string`
>
> Sensitive AWS WAF context from the challenge data.
>
> **website\_url** `string` *required*
>
> Full URL protected by AWS WAF.
>
> **aws\_problem\_url** `string`
>
> Sensitive AWS WAF challenge problem URL, when available.
>
> **wait\_for\_result** `boolean`
>
> Wait for a ready result after creating the task; false returns the task ID immediately.
>
> **aws\_challenge\_js** `string`
>
> Sensitive challenge JavaScript URL or value documented by CapSolver.
>
> **aws\_existing\_token** `string`
>
> Sensitive existing AWS WAF token; never returned.
>
> **wait\_timeout\_seconds** `integer`
>
> Window for scheduling polls. A final scheduled request may use its network timeout; then processing and the task ID are returned.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Solve BotDeflector
>
> **Slug:** `CAPSOLVER_SOLVE_BOTDEFLECTOR`
>
> Create exactly one proxyless BotDeflector task from a target flow token and optionally poll every three seconds for a bounded time. Returns the task ID when processing continues. This is a paid CapSolver operation.
>
> #### Input Parameters
>
> **app\_id** `string`
>
> Optional CapSolver application identifier for usage attribution.
>
> **domain** `string` *required*
>
> Domain used by the target's pow/get or icon/get request. It may differ from website\_url.
>
> **flow\_token** `string` *required*
>
> Fresh target flow token, sometimes named botReflectorJwtToken. It is sent only to CapSolver and never returned.
>
> **website\_url** `string` *required*
>
> Full URL of the page protected by BotDeflector.
>
> **wait\_for\_result** `boolean`
>
> Wait for a ready result after creating the task; false returns the task ID immediately.
>
> **wait\_timeout\_seconds** `integer`
>
> Window for scheduling polls. A final scheduled request may use its three-second network timeout; then processing and the task ID are returned.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Solve Cloudflare Challenge
>
> **Slug:** `CAPSOLVER_SOLVE_CLOUDFLARE_CHALLENGE`
>
> Create one Cloudflare challenge task through a sticky caller-supplied proxy and optionally poll every three seconds for bounded clearance output. This is a paid CapSolver operation.
>
> #### Input Parameters
>
> **html** `string`
>
> Fresh challenge HTML obtained through the same sticky proxy. It is sent upstream and never returned.
>
> **proxy** `string` *required*
>
> Static or sticky caller-controlled proxy in a CapSolver-supported format. Its session must remain stable for at least three minutes and it is never returned.
>
> **app\_id** `string`
>
> CapSolver application identifier for usage attribution.
>
> **user\_agent** `string`
>
> Chrome user agent matching the client and sticky proxy session that will use the clearance result.
>
> **website\_url** `string` *required*
>
> Full URL protected by the Cloudflare challenge.
>
> **wait\_for\_result** `boolean`
>
> Wait for a ready result; false returns the task ID immediately.
>
> **wait\_timeout\_seconds** `integer`
>
> Window for scheduling polls. A final scheduled request may use its three-second network timeout; then processing and the task ID are returned.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Solve Cloudflare Turnstile
>
> **Slug:** `CAPSOLVER_SOLVE_CLOUDFLARE_TURNSTILE`
>
> Create one proxyless Cloudflare Turnstile task and optionally poll every three seconds for a bounded time. Returns the task ID for continuation. This is a paid CapSolver operation.
>
> #### Input Parameters
>
> **cdata** `string`
>
> Secret Turnstile custom data value from widget metadata, when present.
>
> **action** `string`
>
> Turnstile action value from widget metadata, when present.
>
> **app\_id** `string`
>
> CapSolver application identifier for usage attribution, when used.
>
> **website\_key** `string` *required*
>
> Turnstile site key from the target page.
>
> **website\_url** `string` *required*
>
> Full URL of the page containing the Turnstile widget.
>
> **wait\_for\_result** `boolean`
>
> Wait for a ready result; false returns the task ID immediately.
>
> **wait\_timeout\_seconds** `integer`
>
> Window for scheduling polls. A final scheduled request may use its network timeout; then processing and the task ID are returned.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Solve DataDome
>
> **Slug:** `CAPSOLVER_SOLVE_DATADOME`
>
> Create exactly one DataDome slider task with a caller-supplied proxy and optionally wait a bounded time for its solution cookie. This is a paid CapSolver operation.
>
> #### Input Parameters
>
> **html** `string`
>
> Fresh challenge HTML, sent only to CapSolver and never returned.
>
> **proxy** `string` *required*
>
> Caller-controlled proxy in a CapSolver-supported format. Its IP must not be banned by the target, and the value is never returned.
>
> **app\_id** `string`
>
> Optional CapSolver application identifier for usage attribution.
>
> **user\_agent** `string` *required*
>
> Provider-supported browser user agent that exactly matches the client which will use the returned cookie.
>
> **captcha\_url** `string` *required*
>
> Fresh DataDome captcha URL with query parameter t=fe. A URL with t=bv indicates that the proxy IP is banned.
>
> **wait\_for\_result** `boolean`
>
> Wait for a ready result after creating the task; false returns the task ID immediately.
>
> **wait\_timeout\_seconds** `integer`
>
> Window for scheduling polls. A final scheduled request may use its network timeout; then processing and the task ID are returned.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Solve GeeTest
>
> **Slug:** `CAPSOLVER_SOLVE_GEETEST`
>
> Create one GeeTest v3 or v4 token task and optionally poll every three seconds for a bounded time. Returns the task ID when processing continues. This is a paid CapSolver operation.
>
> #### Input Parameters
>
> **app\_id** `string`
>
> Optional CapSolver application identifier for usage attribution.
>
> **version** `string ("v3" | "v4")` *required*
>
> GeeTest protocol version used by the target page.
>
> **challenge** `string`
>
> Fresh GeeTest v3 challenge value; required for version v3.
>
> **risk\_type** `string`
>
> Optional GeeTest v4 risk type when required by the target.
>
> **captcha\_id** `string`
>
> GeeTest v4 captcha identifier; required for version v4.
>
> **website\_url** `string` *required*
>
> Full URL of the page containing GeeTest.
>
> **geetest\_v3\_id** `string`
>
> GeeTest v3 site identifier. Required for version v3; obtain the provider field named gt together with challenge from the target site's fresh GeeTest initialization request. Do not copy a stale value directly from browser developer tools.
>
> **wait\_for\_result** `boolean`
>
> Wait for a ready result after creating the task; false returns the task ID immediately.
>
> **api\_server\_subdomain** `string`
>
> Optional GeeTest API server subdomain used by the target.
>
> **wait\_timeout\_seconds** `integer`
>
> Window for scheduling polls. A final scheduled request may use its three-second network timeout; then processing and the task ID are returned.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Solve Image To Text
>
> **Slug:** `CAPSOLVER_SOLVE_IMAGE_TO_TEXT`
>
> Recognize text or number answers from base64-encoded images and return the immediate recognition result. This is a paid CapSolver operation.
>
> #### Input Parameters
>
> **app\_id** `string`
>
> Optional CapSolver application identifier for usage attribution.
>
> **module** `string ("common" | "number")`
>
> Recognition mode: common returns text; number returns answers for multiple images.
>
> **website\_url** `string`
>
> Optional source page URL associated with the image challenge.
>
> **image\_base64** `string`
>
> Raw base64 image without a data-URI prefix or newlines; required when module is common.
>
> **images\_base64** `array`
>
> One to nine raw base64 images; required when module is number.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Solve MTCaptcha
>
> **Slug:** `CAPSOLVER_SOLVE_MTCAPTCHA`
>
> Create one proxyless or caller-proxied MTCaptcha token task and optionally poll every three seconds for a bounded time. Returns the task ID when processing continues. This is a paid CapSolver operation.
>
> #### Input Parameters
>
> **proxy** `string`
>
> Caller-controlled proxy in a CapSolver-supported format. Required for the proxy variant and never returned.
>
> **app\_id** `string`
>
> CapSolver application identifier for usage attribution.
>
> **variant** `string ("proxyless" | "proxy")`
>
> Whether CapSolver connects directly or through the supplied proxy.
>
> **website\_key** `string` *required*
>
> MTCaptcha site key from the target page.
>
> **website\_url** `string` *required*
>
> Full URL of the page containing MTCaptcha.
>
> **wait\_for\_result** `boolean`
>
> Wait for a ready result; false returns the task ID immediately.
>
> **wait\_timeout\_seconds** `integer`
>
> Window for scheduling polls. A final scheduled request may use its three-second network timeout; then processing and the task ID are returned.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Solve reCAPTCHA V2
>
> **Slug:** `CAPSOLVER_SOLVE_RECAPTCHA_V2`
>
> Solve one standard or enterprise reCAPTCHA v2 challenge in task or synchronous mode. Task mode creates exactly one task and may poll for a bounded time; synchronous mode calls getToken exactly once. This is a paid CapSolver operation.
>
> #### Input Parameters
>
> **proxy** `string`
>
> Caller-controlled proxy in a CapSolver-supported format. Required only for proxy variants and never returned.
>
> **app\_id** `string`
>
> CapSolver application identifier for usage attribution.
>
> **data\_s** `string`
>
> Secret recaptchaDataSValue captured from the target page, when present.
>
> **cookies** `array`
>
> Target cookies for supported sites. Values are sent upstream and never returned.
>
> **variant** `string ("proxyless" | "proxy" | "enterprise_proxyless" | "enterprise_proxy")`
>
> Task variant. Proxy variants require proxy; enterprise variants may use enterprise\_payload.
>
> **api\_domain** `string`
>
> reCAPTCHA API domain used by the target page, when non-default. Common values are [http://www.google.com/](http://www.google.com/) and [http://www.recaptcha.net/](http://www.recaptcha.net/).
>
> **is\_session** `boolean`
>
> Whether CapSolver should request a session-capable solution.
>
> **page\_action** `string`
>
> Action value from the target widget's anchor request, when present.
>
> **website\_key** `string` *required*
>
> reCAPTCHA site key from the target page.
>
> **website\_url** `string` *required*
>
> Full URL of the page containing reCAPTCHA v2.
>
> **is\_invisible** `boolean`
>
> Whether the target uses invisible reCAPTCHA v2.
>
> **execution\_mode** `string ("task" | "synchronous")`
>
> How CapSolver should solve the challenge: task creates exactly one asynchronous task and may poll it; synchronous calls getToken exactly once and returns only a ready result.
>
> **wait\_for\_result** `boolean`
>
> Task mode only: wait for a ready result; false returns the task ID immediately. Ignored in synchronous mode.
>
> **enterprise\_payload** `object`
>
> Secret enterprise payload required by some enterprise integrations.
>
> **wait\_timeout\_seconds** `integer`
>
> Task mode only: window for scheduling polls. A final scheduled request may use its network timeout; then processing and the task ID are returned. Ignored in synchronous mode.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Solve reCAPTCHA V3
>
> **Slug:** `CAPSOLVER_SOLVE_RECAPTCHA_V3`
>
> Solve a standard, enterprise, or M1 reCAPTCHA v3 challenge in task or synchronous mode. Task mode creates exactly one task and may poll for a bounded time. Synchronous mode calls getToken exactly once and rejects M1. M1 variants are high-cost and require provider and site-specific enablement.
>
> #### Input Parameters
>
> **proxy** `string`
>
> Caller-controlled proxy in a CapSolver-supported format. Required only for proxy variants and never returned.
>
> **app\_id** `string`
>
> CapSolver application identifier for usage attribution.
>
> **variant** `string ("proxyless" | "proxy" | "enterprise_proxyless" | "enterprise_proxy" | "m1_proxyless" | "m1_proxy" | "enterprise_m1_proxyless" | "enterprise_m1_proxy")`
>
> reCAPTCHA v3 task variant. Variants ending in \_proxy require proxy. M1 variants are high-cost and require provider and site-specific enablement; use them only when that enablement is confirmed.
>
> **api\_domain** `string`
>
> reCAPTCHA API domain used by the target page, when non-default. Common values are [http://www.google.com/](http://www.google.com/) and [http://www.recaptcha.net/](http://www.recaptcha.net/).
>
> **is\_session** `boolean`
>
> Whether CapSolver should request a session-capable solution.
>
> **page\_action** `string`
>
> Action value whose reCAPTCHA v3 score token is requested.
>
> **website\_key** `string` *required*
>
> reCAPTCHA site key from the target page.
>
> **website\_url** `string` *required*
>
> Full URL of the page containing reCAPTCHA v3.
>
> **execution\_mode** `string ("task" | "synchronous")`
>
> How CapSolver should solve the challenge: task creates exactly one asynchronous task and may poll it; synchronous calls getToken exactly once and returns only a ready result. Synchronous mode does not support M1 variants.
>
> **wait\_for\_result** `boolean`
>
> Task mode only: wait for a ready result; false returns the task ID immediately. Ignored in synchronous mode.
>
> **enterprise\_payload** `object`
>
> Secret enterprise payload required by some target integrations. Accepted only by enterprise variants and never returned.
>
> **wait\_timeout\_seconds** `integer`
>
> Task mode only: window for scheduling polls. A final scheduled request may use its network timeout; then processing and the task ID are returned. Ignored in synchronous mode.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not

> [accordion] Submit Task Feedback
>
> **Slug:** `CAPSOLVER_SUBMIT_TASK_FEEDBACK`
>
> Report whether a previously returned CapSolver token worked, using only the documented minimal task and result fields.
>
> #### Input Parameters
>
> **solved** `boolean` *required*
>
> Whether the returned token successfully solved the target challenge.
>
> **task\_id** `string`
>
> Optional task identifier from the original result.
>
> **task\_type** `string` *required*
>
> Exact CapSolver task type originally submitted, such as ReCaptchaV2TaskProxyLess.
>
> **website\_key** `string`
>
> Optional website key from the original task when that family used one.
>
> **website\_url** `string` *required*
>
> Website URL from the original task; no other original task fields are accepted or returned.
>
> **result\_error\_id** `integer`
>
> Provider errorId from the original result; defaults to zero for a successful result.
>
> #### Output
>
> **data** `string` *required*
>
> Data from the action execution
>
> **error** `string`
>
> Error if any occurred during the execution of the action
>
> **successful** `boolean` *required*
>
> Whether or not the action execution was successful or not
