AI Agents & MCPs & AI Workflow Automation • (~400 MCP servers for AI agents) • AI Automation / AI Agent with MCPs • AI Workflows & AI Agents • MCPs for AI Agents
TypeScript ai-agentai-agent-toolsai-agentsai-agents-frameworkmcpmcp-servermcp-toolsmcpsn8n-alternativeno-code-automationworkflowworkflow-automationworkflows
import { HttpMethod } from '@activepieces/pieces-common';
import { createAction, Property } from '@activepieces/pieces-framework';
import { parallelAuth } from '../auth';
import { parallelClient } from '../common/client';
export const extractAction = createAction({
auth: parallelAuth,
name: 'extract',
classification: 'READ',
displayName: 'Extract Web Content',
description:
'Extract clean, citation-aware content from specific web URLs. Up to 20 URLs per request.',
audience: 'both',
aiMetadata: {
description:
"Fetch and extract clean, citation-aware content from a known list of web URLs (up to 20 per request) via Parallel's research API. Use it when you already have the pages and want their readable content rather than searching for them; an optional objective and queries focus the returned excerpts, and full-page markdown can be requested at the cost of higher latency. Requires at least one URL. Read-only and idempotent.",
idempotent: true,
},
props: {
urls: Property.Array({
displayName: 'URLs',
description: 'List of URLs to extract content from (max 20).',
required: true,
}),
objective: Property.LongText({
displayName: 'Objective',
description:
'Natural-language description of the goal driving the extraction. Used to focus excerpts on the most relevant content.',
required: false,
}),
search_queries: Property.Array({
displayName: 'Search Queries',
description:
'Optional keyword search queries used together with the objective to focus excerpts.',
required: false,
}),
full_content: Property.Checkbox({
displayName: 'Include Full Content',
description:
'When enabled, requests full markdown of each page in addition to excerpts. May increase latency.',
required: false,
defaultValue: false,
}),
max_chars_total: Property.Number({
displayName: 'Max Total Characters',
description: 'Upper bound on total characters across excerpts from all results.',
required: false,
}),
session_id: Property.ShortText({
displayName: 'Session ID',
description: 'Optional session identifier to share context across search/extract calls.',
required: false,
}),
},
async run(context) {
const urls = ((context.propsValue.urls ?? []) as unknown[]).filter(
(u): u is string => typeof u === 'string' && u.trim().length > 0,
);
if (urls.length === 0) {
throw new Error('At least one URL is required.');
}
const queries = ((context.propsValue.search_queries ?? []) as unknown[]).filter(
(q): q is string => typeof q === 'string' && q.trim().length > 0,
);
const body: Record<string, unknown> = { urls };
if (context.propsValue.objective) body['objective'] = context.propsValue.objective;
if (queries.length) body['search_queries'] = queries;
if (context.propsValue.max_chars_total !== undefined && context.propsValue.max_chars_total !== null) {
body['max_chars_total'] = context.propsValue.max_chars_total;
}
if (context.propsValue.session_id) body['session_id'] = context.propsValue.session_id;
if (context.propsValue.full_content) {
body['advanced_settings'] = {
full_content: { enabled: true },
};
}
return await parallelClient.request({
apiKey: context.auth.secret_text,
method: HttpMethod.POST,
path: '/v1/extract',
body,
});
},
});