Parallel offers a Task API for automated, structured web research, transforming natural language queries into precise, schema-compliant outputs.
Parallel offers a Task API for automated, structured web research, transforming natural language queries into precise, schema-compliant outputs. On Nagent, Parallel is exposed as a fully-configurable ai web scraping integration that any agent can call — 32 actions, and API key authentication. No code is required to wire Parallel into your workflow — connect it once via the External Integrations panel and reuse it across every agent you build.
Agent builders use Parallel to automate the kinds of tasks ai web scraping teams previously handled manually. Concrete examples — each one is a single agent step in Nagent — include:
Every action and trigger is paired with a structured input/output schema (visible in the sections below), so when you wire Parallel into Helix — our agentic agent builder — the editor knows exactly what each step expects and produces. Configure once, deploy anywhere across your Nagent agents.
Every operation an agent can call against Parallel, with input parameters and output schema. Drop these into any step of an agent built in Helix.
PARALLEL_ADD_ENRICHMENT_TO_FIND_ALL_RUNTool to add an enrichment to a FindAll run. Use when you need to enrich existing FindAll run results with additional structured data fields. Enrichments define what information to extract from matched entities using a JSON schema.
Input parameters
Processor to use for the enrichment task. Defaults to 'core'.
ID of the FindAll run to add enrichment to.
Optional list of MCP servers to use for the enrichment task.
JSON schema defining the structure and fields for the enrichment output.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PARALLEL_ADD_RUNS_TO_TASK_GROUPTool to initiate multiple task runs within a TaskGroup. Use when you need to execute multiple tasks in parallel within an existing task group.
Input parameters
List of task runs to execute. Up to 1,000 runs can be specified per request.
Unique identifier of the task group to add runs to.
Task specification for defining input and output schemas.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PARALLEL_CANCEL_FIND_ALL_RUNTool to cancel an active FindAll run by findall_id. Use when you need to stop a running FindAll operation before it completes. Cannot cancel runs that have already terminated.
Input parameters
ID of the FindAll run to cancel.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PARALLEL_CREATE_CHAT_COMPLETIONSTool to get realtime chat completions from Parallel AI. Use when you need conversational AI responses or structured outputs via chat interface. Can be combined with Task API processors for research-grade structured outputs with citations and reasoning.
Input parameters
The number of chat completions to generate. Note: This parameter is currently unsupported by the API.
The model to use for chat completions. Available models: 'speed' (low latency), 'lite', 'base', 'core' (research-grade outputs).
The top p to use for chat completions. Note: This parameter is currently unsupported by the API.
Whether to stream the chat completions. When true, returns a stream of partial responses.
The messages to use for chat completions. Must contain at least one message.
The maximum number of tokens to generate. Note: This parameter is currently unsupported by the API.
The temperature to use for chat completions. Note: This parameter is currently unsupported by the API.
Response format configuration for chat completions.
The presence penalty to use for chat completions. Note: This parameter is currently unsupported by the API.
The frequency penalty to use for chat completions. Note: This parameter is currently unsupported by the API.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PARALLEL_CREATE_MONITORTool to create a web monitor that periodically runs the specified query. The monitor runs once at creation and then continues according to the specified cadence (hourly, daily, weekly, or every two weeks). Use when you need to track changes or developments for a specific search query over time.
Input parameters
Search query to monitor for material changes.
Frequency at which the monitor runs: hourly, daily, weekly, or every_two_weeks.
Webhook configuration for monitor notifications.
User-provided metadata stored with the monitor. This field is returned in webhook notifications and GET requests, enabling you to map responses to corresponding objects in your application. For example, if you are building a Slackbot that monitors changes, you could store the Slack thread ID here to properly route webhook responses back to the correct conversation thread.
Output schema configuration for monitor events.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PARALLEL_CREATE_TASK_GROUPTool to create a new task group. Use when batching multiple tasks for parallel execution. Task Groups enable grouping and tracking of multiple task runs within a single manageable unit.
Input parameters
User-provided metadata stored with the task group. Keys are strings, values can be string, integer, number, or boolean.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PARALLEL_CREATE_TASK_RUNTool to create and initiate a task run. Returns immediately with a run object in status 'queued'. Use when you need to execute tasks asynchronously with Parallel AI processors.
Input parameters
Input to the task, either text or a JSON object.
Webhook configuration for task run completion events.
User-provided metadata stored with the run. Keys and values must be strings with a maximum length of 16 and 512 characters respectively.
Processor to use for the task (e.g., 'lite', 'base', 'core', 'pro').
Task specification defining input and output schemas.
Optional list of MCP servers to use for the run. To enable this feature, specify 'mcp-server-2025-07-17' in the 'parallel-beta' header.
Controls tracking of task run execution progress. When set to true, progress events are recorded and can be accessed via the Task Run events endpoint. Note that progress tracking cannot be enabled after a run has been created. The flag is set to true by default for premium processors (pro and above). To enable this feature, specify 'events-sse-2025-07-24' in the 'parallel-beta' header.
Beta features to enable. Comma-separated list of beta feature flags (e.g., 'webhook-2025-08-12,events-sse-2025-07-24'). This will be sent as a header.
Source policy governing preferred and disallowed domains in web search results.
Interaction ID to use as context for this request.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PARALLEL_DELETE_MONITORTool to delete a monitor, stopping all future executions. Use when you need to permanently remove a monitor. Deleted monitors can no longer be updated or retrieved.
Input parameters
Unique identifier of the monitor to delete. Deleted monitors can no longer be updated or retrieved.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PARALLEL_EXTEND_FIND_ALL_RUNTool to extend a FindAll run by adding additional matches to the current match limit. Use when you need to increase the number of matches for an existing FindAll run that is still active or has completed.
Input parameters
ID of the FindAll run to extend.
Additional number of matches to find for this FindAll run. This value will be added to the current match limit to determine the new total match limit. Must be greater than 0.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PARALLEL_EXTRACTTool to extract relevant content from specific web URLs. Use when you need to fetch and extract content from known URLs with optional focusing on specific objectives or search queries.
Input parameters
List of URLs to extract content from
Include excerpts from each URL relevant to the search objective and queries. Can be a boolean or settings object. Note that if neither objective nor search_queries is provided, excerpts are redundant with full content.
If provided, focuses extracted content on the specified search objective
Fetch policy configuration for cache and timeout settings.
Include full content from each URL. Can be a boolean or settings object. Note that if neither objective nor search_queries is provided, excerpts are redundant with full content.
If provided, focuses extracted content on the specified keyword search queries
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PARALLEL_FETCH_TASK_GROUP_RUNSTool to retrieve task runs from a Task Group as a resumable stream. Use when you need to fetch all runs within a group, optionally including their inputs and outputs. The stream can be resumed using the event_id as a cursor.
Input parameters
Filter runs by status (e.g., queued, running, completed, failed, cancelled).
Unique identifier of the Task Group to fetch runs from.
Whether to include task run inputs in the response. Default is False.
ID of the last received event to resume streaming from. Used as cursor for pagination.
Whether to include task run outputs in the response. Default is False.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PARALLEL_FIND_ALLTool to start a FindAll run. Use when you need to discover and match entities based on natural-language objectives. Supports custom conditions, exclusion lists, and webhook callbacks.
Input parameters
Optional webhook configuration for run events.
Optional user metadata to store with the run. Values must be primitive types (string, int, float, or bool).
Quality/cost tier for the run.
Natural-language objective for the FindAll run.
Type of entity to find (e.g., companies, people).
Maximum number of matches to return (between 5 and 1000 inclusive).
Optional list of entities to exclude from matching.
List of conditions that a candidate must satisfy.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PARALLEL_GET_FIND_ALL_RUN_RESULTTool to fetch the final (or latest available) FindAll candidates and result payload for a run. Use when you need to retrieve matched/unmatched candidates after a FindAll run has progressed or completed.
Input parameters
The identifier for the FindAll run.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PARALLEL_GET_FIND_ALL_RUN_SCHEMATool to retrieve the schema configuration of a FindAll run by findall_id. Use when you need to inspect the objective, entity type, match conditions, and other schema details for a previously created FindAll run.
Input parameters
The identifier for the FindAll run.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PARALLEL_INGEST_FIND_ALL_RUNTool to transform a natural language search objective into a structured FindAll specification. Use when you need to generate a FindAll run spec from a user's natural language description. The generated specification serves as a suggested starting point and can be further customized.
Input parameters
Natural language search objective to transform into a structured FindAll specification. Describe what you want to find (e.g., 'Find technology companies in San Francisco with over 100 employees').
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PARALLEL_LIST_MONITOR_EVENTSTool to list events for a monitor from up to the last 300 event groups. Retrieves events including errors and material changes in reverse chronological order.
Input parameters
Unique identifier of the monitor to retrieve events for.
Lookback period to fetch events from. Sample values: `10d`, `1w`. A minimum of 1 day is supported and with one day increments. Use `d` for days, `w` for weeks.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PARALLEL_LIST_MONITORSTool to list active monitors for the user. Returns all monitors regardless of status with their configuration and current state. Supports cursor-based pagination using monitor_id and limit parameters.
Input parameters
Maximum number of monitors to return. If not specified, returns all monitors.
Monitor ID to start listing after (cursor for pagination). Returns monitors starting with this ID in lexicographic order.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PARALLEL_RETRIEVE_EVENT_GROUPTool to retrieve an event group for a monitor. Use when you have a valid monitor ID and event group ID and want to view the execution history.
Input parameters
Unique identifier of the monitor.
Unique identifier of the event group to retrieve.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PARALLEL_RETRIEVE_FIND_ALL_RUN_STATUSTool to retrieve status and metadata for a FindAll run by findall_id. Use when you need to poll or check the progress of a FindAll run that was previously created.
Input parameters
ID of the FindAll run to retrieve.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PARALLEL_RETRIEVE_MONITORTool to retrieve a specific monitor by ID. Returns the monitor configuration including status, cadence, query, and webhook settings.
Input parameters
Unique identifier of the monitor to retrieve.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PARALLEL_RETRIEVE_TASK_GROUPTool to retrieve details of a specific task group. Use when you have a valid task group ID and want to view its details.
Input parameters
Unique identifier of the task group to retrieve.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PARALLEL_RETRIEVE_TASK_GROUP_RUNTool to retrieve run status by run_id for a task group. Use when you need to check the status of a specific task group run or poll for completion.
Input parameters
Unique identifier of the run to retrieve.
Unique identifier of the task group.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PARALLEL_RETRIEVE_TASK_RUNTool to retrieve run status by run_id. Use when you need to check the status or details of a specific task run. The run result is available from the /result endpoint.
Input parameters
ID of the task run to retrieve.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PARALLEL_RETRIEVE_TASK_RUN_INPUTTool to retrieve the input data of a specific task run by run_id. Use when you need to view the original input parameters that were provided to a task run.
Input parameters
Unique identifier of the task run to retrieve input for.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PARALLEL_RETRIEVE_TASK_RUN_RESULTTool to retrieve the result of a task run by run_id, blocking until the run completes. Use when you need to wait for and fetch the final output of a previously initiated task run. The request will block until the run completes or the timeout is reached.
Input parameters
The identifier of the task run to retrieve results for.
Timeout in seconds for blocking request. If not provided, defaults to 600 seconds (10 minutes). The request will block until the run completes or the timeout is reached.
Optional beta version header for accessing beta features.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PARALLEL_SEARCHTool to perform parallel semantic search. Use when you need to retrieve top matching documents for multiple queries in a single call.
Input parameters
List of search query texts to run in parallel Must be non-empty; all entries must be strings.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PARALLEL_SIMULATE_EVENTTool to simulate sending an event for a monitor. Use when testing monitor webhooks or validating monitor configurations. Simulates sending an event of the specified type (defaults to monitor.event.detected).
Input parameters
Event types that can be simulated for a monitor.
Unique identifier of the monitor to simulate an event for.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PARALLEL_STREAM_FIND_ALL_EVENTSTool to stream events from a FindAll run. Use when you need real-time updates on candidate discovery, matching progress, and run status.
Input parameters
Duration in seconds to keep the connection open if no new events occur.
Unique identifier of the FindAll run to stream events from.
ID of the last received event to resume streaming from.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PARALLEL_STREAM_TASK_GROUP_EVENTSTool to stream events for a Task Group. Use when you want real-time updates of group status and run completions.
Input parameters
Duration in seconds to keep the connection open if no new events occur.
Unique identifier of the Task Group to stream events from.
ID of the last received event to resume streaming from.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PARALLEL_STREAM_TASK_RUN_EVENTSTool to stream events for a Task Run. Returns progress updates and state changes for the task run. For runs without enable_events=true, event frequency is reduced.
Input parameters
Unique identifier of the Task Run to stream events from.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PARALLEL_SUGGEST_TASKTool to suggest tasks based on user intent. Use when you need task specifications generated from a natural language description of what you want to accomplish.
Input parameters
User intent to guide the suggestion (e.g., 'Build authentication system for web app')
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
PARALLEL_UPDATE_MONITORTool to update a monitor's configuration. Use when you need to modify an existing monitor's cadence, query, metadata, or webhook settings. At least one field must be non-null to apply an update.
Input parameters
Updated search query for the monitor. Use this for minor updates to prompts and instructions only. Major changes to the query may lead to unexpected results in change detection, as the monitor compares new results with what was previously seen.
Monitor cadence options.
Webhook configuration for monitor updates.
Updated user-provided metadata. This field is returned in webhook notifications, enabling you to map responses to corresponding objects in your application.
Unique identifier of the monitor to update.
Output
Data from the action execution
Error if any occurred during the execution of the action
Whether or not the action execution was successful or not
No publicly available marketplace agent is found using this tool yet. There are 87 agents privately built on Nagent that already use Parallel.
Build on Nagent
Connect Parallel to any Nagent agent in minutes — no API key management, no boilerplate. Just configure and deploy.
The five questions agent builders ask before adopting a new integration.
Open the External Integrations panel inside Nagent (app.nagent.ai/externalIntegration), find Parallel, and click "Connect Now." You'll authenticate with an API key — Nagent handles credential storage and refresh automatically. Once connected, Parallel is available to any agent in your workspace.
No. Nagent provides no-code integration for every tool. Once Parallel is connected, you configure its 32 actions directly in the agent builder UI — no API calls, no boilerplate, no schema management.
Helix — Nagent's agentic agent builder — lets you drop Parallel steps into any workflow visually. Pick an action (e.g., one of those listed above), fill in the inputs (Helix knows the required vs. optional schema for each parameter), and connect it to upstream/downstream steps. Triggers run as the entry point of an agent, so when a Parallel event fires, the agent kicks off automatically.
Every Parallel action and trigger ships with a fully-typed schema — input parameters with name, type, required flag, and description, plus the output payload shape. The schemas are documented in the sections above. Helix uses these schemas to validate your configuration at build time and to type-check the data flowing between steps.
Yes. While Parallel ships with 32 pre-built ai web scraping actions, you can layer custom logic around them inside Helix — pre/post-processing steps, conditional branches, retries, or stitching Parallel together with other connected tools. For deeper customization, talk to our team about Nagent's Agentic AI Lab — forward-deployed engineers who build Parallel-based workflows tailored to your business.