by iliasabk
geolint is a command-line linter that checks a website for AI-search readiness, and its `geolint mcp` command runs an MCP server so assistants can run the audits. It covers AI crawler access, llms.txt, structured data and citability.
PluginContact for pricing
Step 1: Run npx @iliasabk/geolint check yoursite.com for a first report.
Step 2: Review the findings and apply the fix listed for each one.
Step 3: For assistant use, add a geolint server to your MCP client config with command npx and args -y @iliasabk/geolint mcp.
Step 4: Ask the assistant to audit a URL, compare two URLs or generate an llms.txt.
Step 5: Add the GitHub Action or the CLI with --fail-under to CI to catch regressions.
Five read-only tools: audit_url, generate_llms_txt, compare_urls, list_rules and list_ai_bots. They return structured output with per-call timeouts.
No. The quickstart uses npx, so there is no install or config.
No. The README calls it a proposal that no major AI vendor has committed to reading, so those findings are weighted as warnings and hints.
Yes. A GitHub Action produces a score, a SARIF report and a markdown summary, and the CLI exits with code 1 when the score falls below the gate.
Requirements
Deixis MCPDeixis is an MCP server that gives coding agents typed access to Language Server Protocol operations such as definitions, references, diagnostics and guarded renames. It is pre-alpha.
DeviceShelf MCPDeviceShelf MCP lets an AI assistant answer questions about your network from the live inventory and monitoring data in your own DeviceShelf Server. It is local, read-only by default and off until you enable it.
KNX Auditor MCPA design-time KNX and ETS6 assistant exposed as an MCP server. It designs, audits and repairs KNX projects and generates Home Assistant configuration, without access to the live bus.