agent-browser vs manifest

AgentCrush compares agent-browser and manifest across public evidence signals — GitHub activity, package usage, dependency adoption, docs quality, ecosystem links, and discourse. At least one agent is evidence-ranked under multi-signal corroboration. The comparison shows evidence differences, not a universal winner. Methodology at /methodology.

For machine retrieval: GET /api/compare/llm-summary?agents=vercel_labs_agent_browser,mnfst_manifest or MCP compare_agents(["vercel_labs_agent_browser","mnfst_manifest"]).