Crawl & analyse
Status codes, redirects, titles, meta descriptions, canonicals and hreflang: analysis tabs and filters help uncover technical issues.
Open source · Apache-2.0
The crawler for the agentic era.
Understand websites. Find technical SEO issues. Work with people and AI agents. OpenSpider brings crawling and analysis together with open data on your own computer.
macOS · Windows · Linux (CLI)
CRAWL · ANALYSE · CONNECTWhat OpenSpider looks like
OpenSpider 1.0.0-beta.1 on macOS after crawling caloone.de. The images follow your system’s light or dark mode.
- 01 / DISCOVER
Crawl a website
- 02 / UNDERSTAND
Investigate issues
- 03 / TAKE ACTION
Export or query through MCP
One crawl. Many perspectives.
From individual status codes to changes between crawls: findings for your next audit.
Take your results further
CSV and XLSX exports, bulk exports and reports bring crawl results into your wider workflow.
Discover changes
Crawl comparison and Change Detection reveal new, missing and changed URLs, along with changes to page elements.
Automate your workflow
A headless CLI for crawls and exports. Scheduled tasks connect recurring crawls with selected outputs.
Built for agents
The MCP interface gives AI agents access to crawl results. Only read tools are enabled by default.
Local & open source
Crawl data stored as SQLite files on your computer. Apache-2.0. Desktop for macOS and Windows; core and CLI also for Linux x86_64.
MODEL CONTEXT PROTOCOL
For agents
MCP (Model Context Protocol) connects AI agents to tools and data. With OpenSpider, an agent can list saved crawls, query issues and inspect individual URLs. You explicitly enable additional tool classes.
For example: “Show the most important issues in the latest crawl, with relevant URLs.”
list_crawls · list_issues · url_details
BETA
The open beta is live
Try OpenSpider 1.0 for free on macOS or Windows and tell us what you notice. Installers, test tasks and the feedback form are on the beta page.