Manus

Delegates long-running research and analysis tasks to an autonomous AI agent that runs asynchronously with parallel processing. You create a task via API, get back a task_id, then poll until it's done. Works well for market research, stock analysis, product comparisons, or anything requiring multi-source synthesis and report generation. Tasks can take 1-10+ minutes and produce markdown reports, CSVs, and charts. The agent profiles scale from lite (quick lookups) to max (deep research with complex reasoning). Honest take: if you're building something that needs to kick off substantial research without blocking, this handles the async orchestration so you don't have to babysit a long-running process yourself.

Install to Claude Code

npx -y skills add sanjay3290/ai-skills --skill manus --agent claude-code