Skip to content

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

feat(perplexity-ai): ui improvement (#16752)

Talal Jaber1a57aee9a416
diff --git a/bun.lock b/bun.lock
index 8a9a77259a4..767f0456027 100644
--- a/bun.lock
+++ b/bun.lock
@@ -7059,7 +7059,7 @@
},
"packages/pieces/community/perplexity-ai": {
"name": "@activepieces/piece-perplexity-ai",
- "version": "0.4.1",
+ "version": "0.5.0",
"dependencies": {
"@activepieces/core-piece-types": "workspace:*",
"@activepieces/core-utils": "workspace:*",
diff --git a/packages/pieces/community/perplexity-ai/package.json b/packages/pieces/community/perplexity-ai/package.json
index ee668d2c153..fed4a64ffc4 100644
--- a/packages/pieces/community/perplexity-ai/package.json
+++ b/packages/pieces/community/perplexity-ai/package.json
@@ -1,6 +1,6 @@
{
"name": "@activepieces/piece-perplexity-ai",
- "version": "0.4.1",
+ "version": "0.5.0",
"main": "./dist/src/index.js",
"types": "./dist/src/index.d.ts",
"scripts": {
diff --git a/packages/pieces/community/perplexity-ai/src/lib/actions/create-chat-completion.action.ts b/packages/pieces/community/perplexity-ai/src/lib/actions/create-chat-completion.action.ts
index 618634a9f25..9fed75a4c9b 100644
--- a/packages/pieces/community/perplexity-ai/src/lib/actions/create-chat-completion.action.ts
+++ b/packages/pieces/community/perplexity-ai/src/lib/actions/create-chat-completion.action.ts
@@ -22,6 +22,10 @@ export const createChatCompletionAction = createAction({
description:
'Ask a question and get an answer sourced from the live web.',
aiMetadata: { description: 'Sends a question to a Perplexity Sonar model, which answers from a live web search with source citations rather than model memory alone, choosing between the plain sonar models for fast grounded answers and the sonar-reasoning models for step-by-step reasoning. It is the only action in this piece; pick it when the reply must reflect current web content, and prefer a general LLM vendor piece such as OpenAI or Google Gemini when no web lookup is needed. Runs stateless - earlier turns must be supplied in the optional roles array, which alternates user and assistant after any system message; requires a model and a question, and is not idempotent: each call runs a new search and produces a fresh completion.', idempotent: false },
+ propertyGroups: [
+ { key: 'settings', display: 'section', label: 'Settings', icon: 'sliders', props: ['model', 'max_tokens'] },
+ { key: 'message', display: 'section', label: 'Message', icon: 'text', props: ['prompt'] },
+ ],
props: {
model: Property.StaticDropdown({
displayName: 'Model',