Perplexity integration for AI agents.
Perplexity integration for AI agents with secure authentication and server-side credential injection. Open Connector runs the OAuth, seals the token in an encrypted vault, and serves Perplexity tools to your agent over MCP or a typed API — credentials injected server-side, every call audited, nothing leaving your infrastructure. Open source (AGPL-3.0) and self-hostable.
Real Perplexity actions, managed and audited.
Your user connects Perplexity once; your agent can then secure AI-agent actions — scoped to the OAuth permissions you grant and the tool allowlist you configure. Every action is least-privilege and written to a tamper-evident audit trail.
- 1
Your user grants Perplexity access once (OAuth) — the token lands in the vault.
- 2
Your agent calls a tool over MCP or the typed API; Open Connector injects the credential server-side.
- 3
Every routed call appends a hash-chained audit record — nothing leaves your infra.
Supported Perplexity tools.
9 tools are generated from the published Perplexity catalog. Descriptions are plain text; each action remains subject to its configured authentication and tool allowlist.
Showing 9 tools. All published catalog entries are included in this page's server-rendered HTML.
- Create Async Chat Completion
- Queues a Sonar chat completion for asynchronous processing.
- chatasyncresearch
- Create Chat Completion
- Generates a cited, web-grounded Sonar chat completion.
- chatsearchresearch
- Create Contextualized Embeddings
- Creates document-aware embeddings for ordered chunks from the same document.
- embeddingsdocumentsrag
- Create Embeddings
- Creates vector embeddings for independent texts, queries, and sentences.
- embeddingssearchrag
- Execute Agent
- Runs the Perplexity Agent API with a model or a Perplexity preset.
- agentresearchsearch
- Get Async Chat Completion
- Retrieves the current status and, when available, response for an async Sonar request.
- chatasyncread
- List Async Chat Completions
- Lists async Sonar chat completion requests for the connected account.
- chatasyncread
- List Models
- Lists models currently available through the Perplexity Agent API.
- agentmodelsread
- Perplexity Search (Raw Results)
- Returns ranked, real-time web search results without an LLM-generated summary.
- searchwebresearch
Perplexity integration, answered
- How do AI agents use Perplexity through Open Connector?
- Your user connects Perplexity once with one of its cataloged authentication methods. Open Connector stores the credential in an encrypted vault and exposes Perplexity tools to your agent over MCP or a typed API, with credentials injected server-side on each call.
- Is this a Perplexity MCP server?
- Yes. Open Connector can serve Perplexity as a named MCP server with a scoped allowlist and a per-user connection URL, so any MCP client can call Perplexity actions with credentials injected server-side.
- Where do Perplexity credentials live?
- In your own infrastructure. Open Connector keeps credentials in its own vault and injects them at call time, so they never leave your environment.
Give your agents Perplexity — keep the keys.
Open source, self-hostable, with Perplexity credentials that never leave your infrastructure. Run it from source today.