Bring a launch plan into Claude.
Use Claude to inspect launch gaps, work from evidence, and turn one finding into a reviewed change.
Start with a public URL. No account required for the first audit.
Use LaunchBuddy to audit https://example.com. Save the returned fix packs, inspect my authorized repository, and propose one scoped fix.
A finding. A useful resource. A finish line.
Separate launch, technical, and agent-readiness findings. Matching skills and guides. The next three actions, with evidence and a completion check.
Fix one gap, verify it, then learn from a real user.
Give Claude the context.
For Claude Code, add the remote server with this command. In Claude web, use a custom connector if your account supports it. Our public directory listing is a separate review process.
claude mcp add --transport http launchbuddy https://launchbuddy.com/api/mcpAdding the connector does not grant permission to deploy, send messages, or spend.
Try one useful request.
Use LaunchBuddy to audit https://example.com. Save the returned fix packs, inspect my authorized repository, and propose one scoped fix.Use the fix pack to resolve real repository paths before editing. After an authorized deployment, verify the original finding.
Audit the public URL
Use audit_my_launch and retain the returned JSON and native fix packs.
Find the right resource
Use search_launch_resources for a relevant skill, guide, or Swarm launch channel.
Check the result
Verify against the original finding after the authorized change is live. Missing evidence stays unknown.