Actionist·Perplexity
Perplexity

Perplexity

· #279 most-used

Ground every agent decision in live, cited web intelligence

ProductivityAnalyticsDeveloperAIAutomation

Perplexity is an AI-powered search and research platform that delivers grounded, cited answers drawn from live web content. Its Sonar API exposes four model tiers — sonar for fast factual queries, sonar-pro for deep multi-source research, sonar-reasoning for chain-of-thought analysis, and sonar-deep-research for exhaustive multi-step investigation — all returning structured citations alongside every response. Connect Perplexity to Actionist and your agents can run web-grounded research on prospects, competitors, vendors, candidates, and regulations automatically — writing cited findings directly into your CRM, project management tools, and Slack channels without a human ever opening a browser tab.

Try Perplexity with ActionistVisit Perplexity
Average time saved
10 hours
per person · per month
≈ 1 workdays back

Eliminates manual work. Agents eliminate the cycle of manually searching, reading, and summarising web sources before calls, meetings, and decisions — research that previously required a human is completed automatically before the relevant event occurs.

Schedule

What your Perplexity agent runs on autopilot

A week of scheduled jobs your Actionist agent will execute on your behalf.

28Scheduled jobs
7Agents at work
24/7Always on
Agents
Tue–Thu
Tue
Wed
Thu
7a
8a
9a
10a
11a
12p
1p
2p
3p
4p
5p
6p
Multi-app workflows

Perplexity × every other app you use

End-to-end automations that span multiple apps — each one a real business outcome.

6Workflows
6Apps spanned
~24 hrsSaved / week
6Personas served
For sales
Featured4 apps

Prospect brief delivered 10 minutes before every discovery call

When a discovery call is added to the calendar, the Sales Agent retrieves the prospect's company name from HubSpot, runs a Sonar Pro Research task to surface recent strategic news and executive changes, extracts the citations, writes the cited summary to the deal record, and posts a formatted prep brief to the rep's Slack — all within about a minute of the event being created. The rep enters every discovery call with a grounded, timely brief rather than going in cold.

~8 hrs

Time saved for your team — every week, on autopilot

The flow
Trigger·When a discovery call is added to the calendar
Trigger
Step 1
Google Calendar
Detect new discovery call event added to the sales calendar
read
Step 2
HubSpot
Retrieve prospect company name and deal context from CRM
read
Step 3
Perplexity
Run Sonar Pro Research on the prospect — recent news, initiatives, exec changes
read
Step 4
Perplexity
Run Citation Extraction to pull source URLs from the research response
write
Step 5
HubSpot
Write cited research summary to the deal record
write
Step 6
Slack
Post the 3-bullet prep brief with sources to the rep's Slack DM
Result
Write cited research summary to the deal recordPost the 3-bullet prep brief with sources to the rep's Slack DM
The win
Saved per run
25 min
Runs / week
~20×
Reps enter every discovery call with a researched, cited brief
Driven bySales Agent
ROI

Savings

What your team gets back — two angles: what you stop doing manually, and what that's worth.

Without Actionist

What you do manually today

With Actionist

What your agent runs for you

  • Sales
    60 min / week
    Manual pre-call research

    Reps manually google each account before calls, spending 10–20 minutes per prospect piecing together recent news from scattered tabs before typing notes into the CRM.

    Sales Agent
    0 min
    Agent researches every account before every call

    The Sales Agent runs a Perplexity Chat Completion per account on a weekly schedule, writing cited news summaries to HubSpot deal records before Monday calls — reps open the week with pre-researched accounts.

  • Marketing
    75 min / week
    Weekly manual competitive research

    A team member spends 60–90 minutes each Monday visiting competitor sites, reading press releases, and copying notes into a shared doc before the Tuesday planning meeting.

    Marketing Agent
    0 min
    Agent delivers a cited competitive digest before Tuesday planning

    The Marketing Agent runs batch competitor research via Perplexity every Tuesday morning and posts a cited digest to Slack before the team's planning meeting — no dashboard tabs, no manual searches.

  • Customer Support
    45 min / week
    Agent-led research on complex tickets

    Support agents manually search for third-party documentation and community solutions for each complex ticket before composing a reply, adding 10–15 minutes to every complex interaction.

    Customer Support Agent
    0 min
    Agent pre-researches complex tickets before the agent reads them

    When a ticket is tagged complex, the Support Agent queries Perplexity for relevant third-party guides and community solutions, appending cited notes to the ticket before a human agent opens it.

  • Human Resources
    40 min / week
    Manual compensation benchmarking

    HR manually consults salary survey reports, job board ranges, and peer company postings for each role — an hour or more per position, often using data that is 6–12 months out of date.

    Human Resources Agent
    0 min
    Agent surfaces live compensation benchmarks for every open role

    The HR Agent runs a Perplexity Chat Completion per open role each Monday, writing current market salary ranges with citations to the requisition record so offers reflect live market data.

  • Finance
    90 min / week
    Manual due diligence research

    Finance team members manually research each new vendor or acquisition target across news, reviews, and LinkedIn before preparing a due diligence brief — one to two hours per entity.

    Finance Agent
    0 min
    Agent researches every entity before due diligence meetings

    When a new vendor or target enters the pipeline, the Finance Agent triggers a Sonar Pro Research task and writes a cited summary to the due diligence record within about a minute — the team reviews rather than builds from scratch.

  • Operations
    50 min / week
    Manual vendor news monitoring

    Procurement staff manually search for news on key vendors each week — googling company names, scanning LinkedIn, and checking trade press — often missing signals until they become incidents.

    Operations Agent
    0 min
    Agent monitors vendor risk signals every week

    The Operations Agent runs batch Perplexity queries across tier-1 vendors each Wednesday, writes cited risk findings to Notion, and posts elevated-risk vendors to Slack before the procurement review.

  • Legal
    30 min / week
    Manual regulatory monitoring

    Legal team members manually check regulatory body websites and legal publication feeds several times a week, often missing updates that fall between scheduled checks.

    Legal Agent
    0 min
    Agent scans regulatory bodies daily and logs updates with citations

    The Legal Agent runs daily Focused Web Searches on relevant regulatory body sites, logs new guidance or enforcement actions to Notion, and immediately assigns tasks to the responsible counsel.

+ 100s of other Perplexity automations
Average time saved
39 hrs / person / month
Calculator

Calculate what your team saves

Team size
5 people
Hourly rate
$75 / hr
Hours saved / week
13
Hours saved / year
625
Annual ROI
$46,875

Based on Perplexity's typical team usage — the visible tasks plus a few other automations the agent runs: ~2.5 hrs / person / week of admin work automated.

Connect

How to plug Perplexity into Actionist

Pick the connection method that suits your environment.

Connect Actionist to Perplexity using a Sonar API key. Generate one at perplexity.ai/settings/api — the key grants access to all Sonar model endpoints.

1
Open Perplexity API settings

Go to perplexity.ai/settings/api. You need a Perplexity account with API access enabled (available on paid plans).

2
Generate an API key

Click 'Generate' to create a new API key. Copy it immediately — you won't be able to view it again after leaving the page.

3
Paste into Actionist

Paste the key into the API Key field below and click Test connection. Actionist will run a test Chat Completion to confirm the handshake.

Credentials you'll need
API Key*
perplexity.ai/settings/api → Generate a new API key
Actions

12 actions your agent can call

Read and write operations available to your Actionist agent.

Triggers

0 events your agent can react to

Events your agent watches for, and the actions it kicks off in response.

This app has no triggers yet.
Skills

Skills that pair with Perplexity

Reusable agent skills that work well alongside this app.

Perplexity

Search the web with AI-powered answers via Perplexity API. Returns grounded responses with citations. Supports batch queries.

Exa Web Search (Free)

Free AI search via Exa MCP. Web search for news/info, code search for docs/examples from GitHub/StackOverflow, company research for business intel. No API key needed.

Web Search Free

Free AI search via Exa MCP. Fall back to Multi Search Engine if not successful. All without API keys.

MCP servers

MCP servers that work with Perplexity

Connect Actionist to MCP servers built for or around this app.

Perplexity API Platform
Official

Real-time web search, reasoning, and research through Perplexity's API

arjunkmrm-perplexity-search

Enable AI assistants to perform web searches using Perplexity's Sonar Pro.

tanigami/mcp-server-perplexity

Interacting with Perplexity API.

FAQs

Questions about Perplexity + Actionist

How does Actionist connect to Perplexity?
Go to the Apps tab, find Perplexity, and click Connect. The recommended path is API key — navigate to perplexity.ai/settings/api, generate a key, and paste it into the API key field in Actionist. Actionist will run a test call against the Sonar API to confirm the handshake before any agent tasks run. If you prefer MCP, install the Perplexity MCP server from the Actionist MCP catalogue; the server manages the key for you and exposes all Sonar endpoints to any connected agent.
Which Perplexity Sonar models can the agent use?
Actionist agents can call any model available through the Perplexity Sonar API: sonar (fast, cost-efficient web-grounded answers), sonar-pro (deeper multi-source research with more citations), sonar-reasoning (chain-of-thought analytical tasks), and sonar-deep-research (exhaustive multi-step searches that synthesise dozens of sources). The model is selected per action — lightweight fact-lookup steps use sonar; deep competitive research steps use sonar-pro or sonar-deep-research. You can override the default in the action's configuration.
Does the Perplexity API return citations and source URLs?
Yes. Every Sonar API response includes a citations array with the URLs of the web sources used to ground the answer. Actionist agents can extract these citations and pass them downstream — writing them into a Notion page, attaching them to a Google Sheets row, or including them in a Slack digest. Agents do not strip citations; the full provenance of any AI-generated answer is always available in the response payload.
What is the difference between a Chat Completion and a Search action in Actionist?
A Chat Completion action sends a full conversational thread to the Sonar model and returns a web-grounded answer with citations — ideal for agentic research tasks where the agent needs to synthesise an answer from live web content. A Search action is a focused, single-turn query optimised for retrieval speed — the agent submits a keyword or natural-language query and gets back structured results including titles, URLs, and snippets. Use Chat Completion when you need synthesis; use Search when you need a ranked list of sources to act on.
Can I connect Perplexity to other apps in the same agent task?
Yes. Perplexity is most powerful as a research step inside a multi-app agent task. Common patterns: use a Chat Completion to research a prospect, then write the findings to HubSpot; run a Search on a competitor's product and pipe the results into a Google Sheets competitive-tracking sheet; have the agent summarise recent news with Perplexity and post the digest to Slack. Any of Actionist's 200+ connected apps can receive data from — or provide context to — a Perplexity action in the same task.
Does Perplexity's API support streaming responses?
Yes, the Sonar API supports server-sent event streaming. When an Actionist agent uses the Chat Completion action with streaming enabled, tokens arrive incrementally rather than as a single block — this reduces perceived latency for long research responses. Streaming is particularly useful in scheduled agent tasks that generate long-form summaries, because the downstream write step (e.g. posting to Slack or writing to Notion) can begin as soon as the first tokens arrive rather than waiting for the full response.
Does Perplexity have webhooks or event triggers I can react to in Actionist?
Perplexity does not currently offer webhook-based triggers or event subscriptions — it is a request/response API. This means Actionist cannot automatically react to Perplexity events the way it can with apps that emit webhooks. Instead, Perplexity is always the action side of an agent task: another trigger (a schedule, a CRM event, a form submission) fires the task, and the agent calls Perplexity during execution to perform research. Scheduled agent tasks on a cron cadence are the right pattern for recurring research workloads.
How does Perplexity pricing work when agents run many searches?
The Perplexity Sonar API bills per token plus a small per-request fee for web retrieval. The base sonar model starts at $1 per million tokens with a per-request fee of $5–$12 per 1,000 requests depending on context length; sonar-pro runs $3 input / $15 output per million tokens. Actionist does not mark up API costs — you pay Perplexity directly via your API key. For cost control, configure agent tasks to use sonar for lightweight queries and reserve sonar-pro or sonar-deep-research for scheduled research tasks where depth matters.