
Points you at RSS and Atom feeds on any website with clean JSON output. Checks link tags, tests common paths like /feed and /atom, even scans blog subdirectories. Built specifically for programmatic use, so stdout is pure JSON with no chatty logs unless you want them. Exit codes tell you if feeds were found, and it validates that URLs actually return XML instead of 404s. The timeout normalization and diagnostics array are nice touches if you're building feed aggregators or monitoring tools. Faster than manually parsing HTML, and the parallel processing means you can check multiple sites without waiting around.
npx -y skills add brooksy4503/rss-agent-discovery --skill rss-agent-discovery --agent claude-codeInstalls into .claude/skills of the current project.
Select a file.