searxng-mcp-server

Provides an MCP server that exposes SearxNG search capabilities (web, images, videos, news) plus a fetch_url utility that retrieves a URL and converts it to markdown. Can run over stdio (default) or HTTP (via MCP_PORT/--port) and is configurable with a required SEARXNG_URL and optional timeout/user-agent/log level.

Evaluated Apr 04, 2026 (25d ago)
Homepage ↗ Repo ↗ Search mcp searxng search web-scraping tooling python
⚙ Agent Friendliness
45
/ 100
Can an agent use this?
🔒 Security
35
/ 100
Is it safe for agents?
⚡ Reliability
18
/ 100
Does it work consistently?

Score Breakdown

⚙ Agent Friendliness

MCP Quality
75
Documentation
70
Error Messages
0
Auth Simplicity
20
Rate Limits
10

🔒 Security

TLS Enforcement
60
Auth Strength
10
Scope Granularity
0
Dep. Hygiene
55
Secret Handling
60

TLS is not discussed; HTTP transport is described with http://0.0.0.0:<port>. No authentication/authorization is documented for the MCP server. The tool includes fetch_url that can retrieve and convert remote content; this can be risky with untrusted inputs (SSRF-like concerns depending on implementation). Dependency set includes common libraries (fastmcp, httpx, pydantic, markitdown) but no security posture/CVE status is provided.

⚡ Reliability

Uptime/SLA
0
Version Stability
40
Breaking Changes
0
Error Recovery
30
AF Security Reliability

Best When

You already operate a SearxNG instance and want an MCP-compatible bridge to use it from an agent, especially over stdio.

Avoid When

You cannot trust inputs to fetch_url (e.g., user-provided arbitrary URLs) or you need an authenticated/permissioned MCP endpoint for untrusted clients.

Use Cases

  • Add privacy-respecting metasearch as a tool in an AI agent via MCP
  • Perform web/image/video/news discovery with optional language and time filtering
  • Fetch and summarize/transform web pages into markdown for downstream LLM processing
  • Run search tools locally (agent uses MCP over stdio) while pointing at an existing SearxNG instance

Not For

  • Production deployments that require strong authentication/authorization on the MCP server itself (none is documented)
  • Use cases that need strict data residency guarantees or formal compliance statements
  • Workloads that require documented, predictable pagination/result limits and strong rate-limit handling guarantees
  • Environments where running arbitrary URL fetching (fetch_url) would be unsafe

Interface

REST API
No
GraphQL
No
gRPC
No
MCP Server
Yes
SDK
No
Webhooks
No

Authentication

OAuth: No Scopes: No

No authentication/authorization mechanism for the MCP server is described. Access control for HTTP transport (if used) is therefore unclear from the provided README.

Pricing

Free tier: No
Requires CC: No

No pricing information is provided; the package appears to be an installable open-source component.

Agent Metadata

Pagination
none
Idempotent
False
Retry Guidance
Not documented

Known Gotchas

  • fetch_url can introduce latency and failure modes depending on remote sites; retry behavior is not documented
  • When using HTTP transport, the server listens on 0.0.0.0 without documented authentication; avoid exposing to untrusted networks
  • No explicit rate-limit guidance is documented; underlying SearxNG behavior may vary by deployment

Alternatives

Full Evaluation Report

Comprehensive deep-dive: security analysis, reliability audit, agent experience review, cost modeling, competitive positioning, and improvement roadmap for searxng-mcp-server.

AI-powered analysis · PDF + markdown · Delivered within 30 minutes

$99

Package Brief

Quick verdict, integration guide, cost projections, gotchas with workarounds, and alternatives comparison.

Delivered within 10 minutes

$3

Score Monitoring

Get alerted when this package's AF, security, or reliability scores change significantly. Stay ahead of regressions.

Continuous monitoring

$3/mo

Scores are editorial opinions as of 2026-04-04.

8642
Packages Evaluated
17761
Need Evaluation
586
Need Re-evaluation
Community Powered