Skip to main content
Glama
figranium

Figranium MCP Server

Official

create_task

Build complete browser automation workflows with sequential actions, state variables, anti-bot stealth, and optional scheduling for recurring data extraction and web testing.

Instructions

Create a complete, fully-configured Figranium automation task including sequential action steps, state variables, anti-bot stealth mechanisms, and optional scheduling.

!!! IMPORTANT GUIDELINES FOR LLM AGENTS !!!

  1. VARIABLE TEMPLATING SYNTAX: You MUST use {$variable_name} (with a single curly brace and dollar sign, e.g. {$myVar}) for variable references/templating inside action values, URLs, headers, or body fields. NEVER use double curly braces like {{variable_name}} or JavaScript-style templates like ${variable_name}, as these syntaxes are unsupported and will cause execution failures.

  2. TASK VALIDATION: Newly created or updated tasks are ordinarily intended to be validated against a real execution result so configuration, selectors, and extracted output can be checked.

  3. MODE SELECTION: Use agent mode by default, including for scraping tasks. scrape mode does not support action blocks and should be used only when extremely fast, action-free scraping is required. Use headful for visible interactive debugging.

1. Purpose

Use this tool when you need to automate any recurring or complex web-based workflows, including data extraction (scraping), automated form-filling, dashboard testing, or dynamic visual monitoring. Tasks are stored permanently in Figranium and can be executed ad-hoc, triggered via API, or scheduled.

2. Execution Model

Figranium tasks run as a linear sequence of steps defined in the 'actions' array. Actions are processed in order from top to bottom. Control flow steps (such as 'if', 'while', 'repeat') allow loops and branching, while 'on_error' steps define fallback behaviors. Variables represent the state and can be updated dynamically during execution. Ensure all opened block structures (such as 'if', 'while', 'repeat', 'foreach') are closed with an 'end' action step.

3. Comprehensive Step Types

  • 'navigate': Redirect browser to a new URL specified in the 'value' field.

  • 'wait': Pause execution for N seconds specified in the 'value' field.

  • 'wait_selector': Pause until the DOM element matching 'selector' is rendered.

  • 'click': Simulate a single, double, or right click on the element matching 'selector'; set 'clickType' to 'double' or 'right' when needed.

  • 'check' / 'uncheck': Idempotently set the checked state of a checkbox or radio control matching 'selector'.

  • 'drag_and_drop': Drag from 'selector' to the required 'targetSelector'.

  • 'reload': Reload the current page and wait for DOM content to load.

  • 'select': Choose an option from a native select using 'selector' and its option 'value'.

  • 'type': Type the 'value' into the 'selector' input element. Use 'typeMode' to clear/replace or append.

  • 'hover': Move mouse pointer to the element matching 'selector'.

  • 'press': Press a specific keyboard key (e.g., 'Enter') specified in the 'key' field.

  • 'scroll': Scroll the page or target element to a specific coordinate or direction.

  • 'javascript': Execute custom JavaScript on the page. Stored in 'value', outputs can be saved to 'varName'.

  • 'screenshot': Capture and save a screenshot.

  • 'http_request': Perform direct API requests.

  • 'if', 'else', 'end': Conditional blocks based on variables.

  • 'while', 'repeat', 'foreach': Looping blocks.

  • 'stop': Halt task execution.

  • 'set': Set or update a task variable.

  • 'solve_captcha': Attempt to automatically solve a detected CAPTCHA challenge.

  • 'wait_captcha': Pause until a CAPTCHA challenge is initialized/ready without solving it.

  • 'upload': Attach the newest unuploaded file, ZIP, or folder from a Cabinet (see 'cabinetId') to a file input, chooser, or drop target matching 'selector'.

  • 'finalize_uploads': Mark all Cabinet items attached during the execution as uploaded.

4. Selector Strategy & Fallbacks

When targeting elements, follow this hierarchy of selectors:

  1. Unique IDs (e.g., '#submit-button')

  2. ARIA roles and labels (e.g., '[aria-label="Search"]')

  3. Reliable CSS classes or data attributes (e.g., '.btn-primary', '[data-testid="login"]')

  4. Text matchers or XPath as a final resort. Fallback: If an element might be missing or slow to load, wrap the interaction inside an 'if' block evaluating a variable or use 'on_error' to catch failure.

5. Edge Cases & Retry Logic

  • Timeouts: Wait-selectors have a default timeout. Ensure critical steps use 'wait_selector' first to avoid clicking non-existent elements.

  • Stealth: Turning on options like 'naturalTyping', 'cursorGlide', and 'allowTypos' simulates authentic human speed and rhythm to prevent anti-bot blocking on protected sites.

  • Statelessness: Enable 'statelessExecution' to ensure execution is completely fresh without persistent browser storage/cookies.

6. Complex Real-World Multi-Step JSON Example:

{
  "name": "HackerNews Custom Scraper",
  "url": "https://news.ycombinator.com",
  "mode": "agent",
  "wait": 3,
  "rotateUserAgents": true,
  "stealth": {
    "allowTypos": true,
    "cursorGlide": true,
    "naturalTyping": true
  },
  "actions": [
    {
      "type": "wait_selector",
      "selector": ".hnname"
    },
    {
      "type": "click",
      "selector": "a.hnmore"
    },
    {
      "type": "wait",
      "value": "2"
    },
    {
      "type": "javascript",
      "value": "return Array.from(document.querySelectorAll('.athing')).map(tr => ({ id: tr.id, title: tr.querySelector('.titleline > a')?.innerText, href: tr.querySelector('.titleline > a')?.href }));",
      "varName": "hn_stories"
    },
    {
      "type": "navigate",
      "value": "https://httpbin.org/post"
    },
    {
      "type": "wait_selector",
      "selector": "pre"
    },
    {
      "type": "javascript",
      "value": "console.log('Finished scraping and navigated successfully.');"
    }
  ],
  "variables": {},
  "extractionFormat": "json"
}

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
urlYesInitial URL to navigate to when the task starts. Expected type: string. Example: 'https://news.ycombinator.com'
modeYesExecution mode. Use 'agent' by default, including for scraping, because it supports action blocks. 'scrape' does not support action blocks and is only for exceptional, extremely fast action-free scraping. 'headful' runs in a visible browser window with human oversight. Expected type: string enum. Example: 'agent'
nameYesDescriptive name of the automation task. Expected type: string. Example: 'HackerNews Scraper'
waitNoStandard delay in seconds to wait after navigation and page loads to let dynamic scripts complete. Expected type: number. Example: 5
actionsNoSequential list of browser actions/control flow steps to execute. Action blocks require 'agent' or 'headful' mode and are not supported in 'scrape' mode. Note: Variable references MUST use '{$variable_name}' syntax.
stealthNoConfigures realistic stealth, anti-bot, and human behavior simulation on the browser instance.
scheduleNoTask automatic execution schedule. Expected type: object.
selectorNoDefault CSS selector to wait for on the page load before starting actions. Expected type: string. Example: '.main-content'
variablesNoTask INPUT variables and caller-overridable configuration only. Do not declare output-only fields or transient runtime state here. Use Set Variable for mid-task state. Expected type: record object of variable configurations.
descriptionNoDetailed description of what the task automates. Expected type: string. Example: 'Logs in and extracts weekly leads'
humanTypingNoVary typing speeds and insert tiny delays to simulate organic human typing. Expected type: boolean. Example: true
includeHtmlNoWhether to include the raw page HTML in the execution response. Expected type: boolean. Example: false
translationNoOptional page translation for Agent and headful runs. It is disabled by default and is not available in Scrape mode.
rotateProxiesNoRotate through configured proxy IPs to prevent IP-based rate limiting. Expected type: boolean. Example: false
rotateViewportNoVary viewport resolutions randomly to simulate multiple desktop and mobile devices. Expected type: boolean. Example: true
disableRecordingNoDisable video/VNC recording of this task to save storage. Expected type: boolean. Example: true
extractionFormatNoTarget export format of any extracted data. Expected type: string enum. Example: 'json'json
extractionScriptNoPost-execution extraction script for the task's actual final output. Final structured fields, lists, tables, parsing, and extraction logic belong here rather than in task variables or JavaScript action blocks. Expected type: string. Example: 'return Array.from(document.querySelectorAll("a")).map(el => el.href)'
includeShadowDomNoWhether to parse and resolve target elements residing in Shadow DOMs. Expected type: boolean. Example: true
rotateUserAgentsNoRotate user agents across requests to avoid pattern blocking and fingerprinting. Expected type: boolean. Example: true
downloadCabinetIdNoCabinet used for intercepted downloads; omitted uses the default Cabinet. Expected type: string. Example: 'cab_basic'
statelessExecutionNoIf set to true, clear browser cookies and session states between runs. Expected type: boolean. Example: false

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed6 schema fields changedv1.3.0
    • addedInput schema / properties / actions / items / properties / clickType
      Added value: +{
      +  "default": "single",
      +  "description": "Click interaction mode for 'click': single, double, or right click. Expected type: string enum. Example: 'double'",
      +  "enum": [
      +    "single",
      +    "double",
      +    "right"
      +  ],
      +  "type": "string"
      +}
    • addedInput schema / properties / actions / items / properties / targetSelector
      Added value: +{
      +  "description": "Destination selector for 'drag_and_drop'. Required with a source 'selector'. Expected type: string. Example: '.done-column'",
      +  "type": "string"
      +}
    • changedInput schema / properties / actions / items / properties / type / enum
      Previous value: -[
      -  "click",
      -  "type",
      -  "wait",
      -  "wait_selector",
      -  "press",
      -  "scroll",
      -  "javascript",
      -  "csv",
      -  "hover",
      -  "merge",
      -  "screenshot",
      -  "if",
      -  "else",
      -  "end",
      -  "while",
      -  "repeat",
      -  "foreach",
      -  "stop",
      -  "set",
      -  "on_error",
      -  "navigate",
      -  "wait_downloads",
      -  "start",
      -  "http_request",
      -  "get_content",
      -  "solve_captcha",
      -  "wait_captcha",
      -  "upload",
      -  "finalize_uploads"
      -]New value: +[
      +  "click",
      +  "type",
      +  "wait",
      +  "wait_selector",
      +  "press",
      +  "scroll",
      +  "javascript",
      +  "csv",
      +  "hover",
      +  "merge",
      +  "screenshot",
      +  "if",
      +  "else",
      +  "end",
      +  "while",
      +  "repeat",
      +  "foreach",
      +  "stop",
      +  "set",
      +  "on_error",
      +  "navigate",
      +  "wait_downloads",
      +  "start",
      +  "http_request",
      +  "get_content",
      +  "solve_captcha",
      +  "wait_captcha",
      +  "upload",
      +  "finalize_uploads",
      +  "check",
      +  "uncheck",
      +  "drag_and_drop",
      +  "reload",
      +  "select",
      +  "do_nothing"
      +]
    • removedInput schema / properties / cabinetId
      Removed value: -{
      -  "description": "Cabinet used for intercepted downloads and 'upload' actions that omit their own cabinetId; omitted uses the default Cabinet. Expected type: string. Example: 'cab_basic'",
      -  "type": "string"
      -}
    • addedInput schema / properties / downloadCabinetId
      Added value: +{
      +  "description": "Cabinet used for intercepted downloads; omitted uses the default Cabinet. Expected type: string. Example: 'cab_basic'",
      +  "type": "string"
      +}
    • addedInput schema / properties / translation
      Added value: +{
      +  "description": "Optional page translation for Agent and headful runs. It is disabled by default and is not available in Scrape mode.",
      +  "properties": {
      +    "enabled": {
      +      "description": "Enable rendered-page translation. Expected type: boolean. Example: true",
      +      "type": "boolean"
      +    },
      +    "targetLanguage": {
      +      "description": "translate.js target language name. Expected type: string. Example: 'spanish'",
      +      "type": "string"
      +    }
      +  },
      +  "required": [
      +    "enabled",
      +    "targetLanguage"
      +  ],
      +  "type": "object"
      +}
  2. Changed4 schema fields changedv1.2.3
    • changedInput schema / properties / actions / items / properties / varName / description
      Previous value: -"Variable name to store output data or extracted content in. Expected type: string. Example: 'extractedTitle'"New value: +"Runtime variable name used by actions such as Set Variable, merge, or foreach. Set Variable may create or update this runtime variable. Do not use varName as the schema for final task outputs; use extractionScript for final extraction."
    • changedInput schema / properties / extractionScript / description
      Previous value: -"Optional post-execution script to extract data. Expected type: string. Example: 'return Array.from(document.querySelectorAll(\"a\")).map(el => el.href)'"New value: +"Post-execution extraction script for the task's actual final output. Final structured fields, lists, tables, parsing, and extraction logic belong here rather than in task variables or JavaScript action blocks. Expected type: string. Example: 'return Array.from(document.querySelectorAll(\"a\")).map(el => el.href)'"
    • changedInput schema / properties / variables / additionalProperties / description
      Previous value: -"Configures state variables accessible throughout the task execution."New value: +"Configures task input variables and caller-overridable values. Do not declare output-only fields here; use Set Variable for mid-task runtime state and extractionScript for final outputs."
    • changedInput schema / properties / variables / description
      Previous value: -"Task variables to store state and dynamic values. Expected type: record object of variable configurations."New value: +"Task INPUT variables and caller-overridable configuration only. Do not declare output-only fields or transient runtime state here. Use Set Variable for mid-task state. Expected type: record object of variable configurations."
  3. Changed8 schema fields changedv1.2.1
    • changedInput schema / properties / actions / description
      Previous value: -"Sequential list of browser actions/control flow steps to execute. Note: Variable reference MUST use '{$variable_name}' syntax."New value: +"Sequential list of browser actions/control flow steps to execute. Action blocks require 'agent' or 'headful' mode and are not supported in 'scrape' mode. Note: Variable references MUST use '{$variable_name}' syntax."
    • addedInput schema / properties / actions / items / properties / cabinetId
      Added value: +{
      +  "description": "Source Cabinet ID for an 'upload' action; omitted uses the default Cabinet. Expected type: string. Example: 'cab_basic'",
      +  "type": "string"
      +}
    • addedInput schema / properties / actions / items / properties / captchaType
      Added value: +{
      +  "description": "The CAPTCHA provider to target for 'solve_captcha'/'wait_captcha' actions. Expected type: string enum. Example: 'recaptcha_v2'",
      +  "enum": [
      +    "recaptcha_v2",
      +    "recaptcha_v3",
      +    "hcaptcha",
      +    "turnstile"
      +  ],
      +  "type": "string"
      +}
    • addedInput schema / properties / actions / items / properties / markAsUploaded
      Added value: +{
      +  "default": false,
      +  "description": "When true, an 'upload' action marks its Cabinet item uploaded immediately after attaching it. Expected type: boolean. Example: false",
      +  "type": "boolean"
      +}
    • addedInput schema / properties / actions / items / properties / timeout
      Added value: +{
      +  "description": "Maximum time in seconds to wait for a CAPTCHA to become ready or solved, for 'solve_captcha'/'wait_captcha' actions. Expected type: number. Example: 30",
      +  "type": "number"
      +}
    • changedInput schema / properties / actions / items / properties / type / enum
      Previous value: -[
      -  "click",
      -  "type",
      -  "wait",
      -  "wait_selector",
      -  "press",
      -  "scroll",
      -  "javascript",
      -  "csv",
      -  "hover",
      -  "merge",
      -  "screenshot",
      -  "if",
      -  "else",
      -  "end",
      -  "while",
      -  "repeat",
      -  "foreach",
      -  "stop",
      -  "set",
      -  "on_error",
      -  "navigate",
      -  "wait_downloads",
      -  "start",
      -  "http_request",
      -  "get_content"
      -]New value: +[
      +  "click",
      +  "type",
      +  "wait",
      +  "wait_selector",
      +  "press",
      +  "scroll",
      +  "javascript",
      +  "csv",
      +  "hover",
      +  "merge",
      +  "screenshot",
      +  "if",
      +  "else",
      +  "end",
      +  "while",
      +  "repeat",
      +  "foreach",
      +  "stop",
      +  "set",
      +  "on_error",
      +  "navigate",
      +  "wait_downloads",
      +  "start",
      +  "http_request",
      +  "get_content",
      +  "solve_captcha",
      +  "wait_captcha",
      +  "upload",
      +  "finalize_uploads"
      +]
    • addedInput schema / properties / cabinetId
      Added value: +{
      +  "description": "Cabinet used for intercepted downloads and 'upload' actions that omit their own cabinetId; omitted uses the default Cabinet. Expected type: string. Example: 'cab_basic'",
      +  "type": "string"
      +}
    • changedInput schema / properties / mode / description
      Previous value: -"Execution mode. 'scrape' is fast and headless; 'agent' uses automated browser interaction; 'headful' runs in a visible browser window with human oversight. Expected type: string enum. Example: 'agent'"New value: +"Execution mode. Use 'agent' by default, including for scraping, because it supports action blocks. 'scrape' does not support action blocks and is only for exceptional, extremely fast action-free scraping. 'headful' runs in a visible browser window with human oversight. Expected type: string enum. Example: 'agent'"
  4. Changed5 schema fields changedv1.1.1
    • changedInput schema / properties / actions / description
      Previous value: -"Sequential list of browser actions/control flow steps to execute."New value: +"Sequential list of browser actions/control flow steps to execute. Note: Variable reference MUST use '{$variable_name}' syntax."
    • changedInput schema / properties / actions / items / description
      Previous value: -"Represents a discrete automation step or flow-control operation executed in sequence."New value: +"Represents a discrete automation step or flow-control operation executed in sequence. Variable reference MUST use '{$variable_name}' syntax."
    • changedInput schema / properties / actions / items / properties / body / description
      Previous value: -"Payload body for 'http_request' actions. Expected type: string. Example: '{\"query\": \"sales\"}'"New value: +"Payload body for 'http_request' actions. Supports variable templating. MUST use '{$variable_name}' syntax for variable references (e.g., '{\"query\": \"{$value}\"}'). NEVER use '{{variable_name}}' or '${variable_name}'. Expected type: string. Example: '{\"query\": \"{$value}\"}'"
    • changedInput schema / properties / actions / items / properties / headers / description
      Previous value: -"JSON stringified headers for 'http_request'. Expected type: string. Example: '{\"Authorization\": \"Bearer x\"}'"New value: +"JSON stringified headers for 'http_request'. Supports variable templating. MUST use '{$variable_name}' syntax for variable references (e.g., '{\"Authorization\": \"Bearer {$token}\"}'). NEVER use '{{variable_name}}' or '${variable_name}'. Expected type: string. Example: '{\"Authorization\": \"Bearer {$token}\"}'"
    • changedInput schema / properties / actions / items / properties / value / description
      Previous value: -"Input value or configuration value for this action. Used for typing text, script contents, or wait durations. Expected type: string. Example: 'hello@world.com'"New value: +"Input value or configuration value for this action. Supports variable templating. MUST use '{$variable_name}' syntax for variable references (e.g., '{$myVar}'). NEVER use '{{variable_name}}' or '${variable_name}'. Expected type: string. Example: 'hello@world.com'"
  5. First observedv1.0.0

TDQS

A4.3/5.0
Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Beyond the annotations, it discloses persistence ('stored permanently'), the execution model, the intended post-creation validation against real results, and hard failure constraints around variable templating. It does not mention rate limits or auth requirements, but the annotations already carry the basic read/write and destructiveness profile.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is long, but the critical variable-syntax and mode rules are front-loaded and the rest is organized by headings with a concrete JSON example. Slight repetition of the templating rule across multiple fields is intentional and earns its place in a complex schema.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a 22-parameter task-creation tool, it covers purpose, execution model, all action types, selector strategy, edge cases, and provides a full example. It does not describe the tool's return value or response shape, but the creation semantics and boundaries are otherwise complete.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

The schema already documents all 22 parameters, but the description adds real semantics: what each action type does, selector precedence, when to use wait_selector, and the mandatory {$var} templating syntax. It goes beyond the schema without having to carry the whole load.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The opening line states a specific action and object: 'Create a complete, fully-configured Figranium automation task' and enumerates its components (actions, variables, stealth, scheduling). It clearly differentiates create_task from task_update/task_delete and gives purpose context in section 1.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Section 1 explicitly says 'Use this tool when you need to automate...' and the mode-selection guidance tells when agent/scrape/headful are appropriate. However, it does not explicitly name sibling tools or say when NOT to use create_task versus task_update/task_execute.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.