Perplexity AI
for LuumenAI
Web-grounded answers and long-running research jobs from the terminal
Connect Perplexity AI and the agent can search the web and ask Sonar models for sourced answers while it works a ticket. Useful for checking a CVE, reading a vendor advisory, or confirming an end-of-support date without leaving the terminal. Longer research runs as an async job you approve first, then collect when it finishes. Luumen authenticates with an API key.
The Perplexity AI toolbox
9 tools: 7 read, 2 write. Reads answer instantly. Writes require approval by default. Everything is logged.
- ReadCreate Chat CompletionPerplexity Sonar Chat Completions (POST /v1/sonar).
- ReadCreate Contextualized EmbeddingsCreate Contextualized Embeddings (POST /v1/contextualizedembeddings).
- ReadCreateEmbeddingsGenerate vector embeddings for independent texts (queries, sentences, documents).
- ReadGet Async Chat CompletionGet Async Chat Completion (GET /v1/async/sonar/{id}).
- ReadList Async Chat CompletionsList Async Chat Completions (GET /v1/async/sonar).
- ReadList ModelsList Models (GET /v1/models).
- ReadPerplexity Search (Raw Results)Search the Web (POST /search).
- WriteCreate Async Chat CompletionCreate Async Chat Completion (POST /v1/async/sonar). Approval by default
- WriteExecute AgentCreate Agent Response (POST /v1/agent). Approval by default
One prompt, start to finish
What a governed Perplexity AI run looks like inside Luumen.
Questions
Authorize once with API token. Luumen lists the scopes each action needs before you approve the connection, and credentials never appear in the chat.
Read actions answer immediately. Anything that writes — create async chat completion, execute agent — is shown as a plan and requires approval by default. Administrators configure that per tool, so you decide exactly which actions can ever run unattended.
You decide. Actions are granted per agent, skill, and team, and per environment — production is not staging. Read access can be broad while writes stay narrow.
Every call to Perplexity AI — read or write, approved or declined — is recorded with the actor, the input, and the result, and can be linked to the ticket or change record.
Often connected alongside
All integrations
Adobe Marketing Agent MCPAdobe Experience Platform context, sandbox switches, and async task status
AffindaRead parsed documents, collections, and annotations — change them with approval
Agent MailGive the agent its own mailbox for reading threads and sending updates
AI/ML APIInspect assistants, threads, and runs — and change them as approved stepsPut Perplexity AI to work with Luumen
Connect in minutes. Every action scoped, approved, and audited from day one.