perplexica-mcp
Provides an MCP server (via FastMCP) that exposes a Perplexica-backed search tool to MCP clients. Supports multiple MCP transport modes: stdio, SSE, and Streamable HTTP (with Docker deployment options).
Score Breakdown
⚙ Agent Friendliness
🔒 Security
Uses HTTPS/TLS is not explicitly documented; deployments may expose SSE/HTTP endpoints without guidance on TLS. No authentication/authorization for the MCP server is described (likely relies on network controls and the Perplexica backend). Environment variables are used for backend URL and model provider names; no explicit guidance on secret storage/logging is provided in the excerpt. Dependencies include fastmcp, httpx, and python-dotenv, but no vulnerability status is provided.
⚡ Reliability
Best When
You want a drop-in MCP server that bridges MCP clients to a self-hosted Perplexica search backend using stdio (most compatible) or SSE/HTTP for networked use.
Avoid When
You need strict rate-limit guarantees, strong end-user auth, or documented error codes/retry semantics at the MCP tool level (not evidenced in the provided content).
Use Cases
- • Enable MCP-compatible IDE/desktop tools to perform Perplexica search from within workflows
- • Web/academic/discussion search via a unified MCP tool interface
- • Self-hosted deployment of Perplexica search accessible to various MCP clients
- • Prototype and production use with containerized SSE/HTTP transports
Not For
- • Use cases that require a dedicated/authenticated API surface with OAuth for the MCP server itself (it appears to rely on Perplexica backend configuration)
- • Environments that cannot provide the required Perplexica backend URL and model/provider env vars
- • Apps needing first-class pagination or cursor-style result APIs (not described)
Interface
Authentication
No authentication mechanism for the MCP server is described. Environment variables are used to connect to the Perplexica backend and configure model providers.
Pricing
Pricing not mentioned; likely costs depend on Perplexica backend and underlying model providers (e.g., OpenAI/others) configured via environment variables.
Agent Metadata
Known Gotchas
- ⚠ Misconfiguration risk: required PERPLEXICA_BACKEND_URL and model provider/name env vars may not be set, causing connection/model initialization failures.
- ⚠ Transport mismatch: some MCP clients expect stdio configuration; SSE/HTTP require separate server startup and different client settings.
- ⚠ No documented retry/idempotency guidance at the tool level; agents may need to implement safe retries themselves if desired.
- ⚠ As a bridge to Perplexica, runtime errors may originate from Perplexica/backend dependencies rather than the MCP wrapper.
Alternatives
Full Evaluation Report
Comprehensive deep-dive: security analysis, reliability audit, agent experience review, cost modeling, competitive positioning, and improvement roadmap for perplexica-mcp.
AI-powered analysis · PDF + markdown · Delivered within 30 minutes
Package Brief
Quick verdict, integration guide, cost projections, gotchas with workarounds, and alternatives comparison.
Delivered within 10 minutes
Score Monitoring
Get alerted when this package's AF, security, or reliability scores change significantly. Stay ahead of regressions.
Continuous monitoring
Scores are editorial opinions as of 2026-03-30.