setup-check validates Python version, package install, API keys, dependencies, and workspace configuration. The checks are local: they confirm a key is set, not that the provider accepts it. check-login only writes a local marker file and always reports success.
swarms agent \ --name "Research Assistant" \ --description "AI research specialist" \ --system-prompt "You are an expert research assistant. Provide structured, evidence-based answers." \ --task "Summarize key quantum computing breakthroughs from the last two years" \ --model-name "claude-sonnet-4-6" \ --temperature 0.1 \ --max-loops 3
Each .md file uses frontmatter for name, description, model_name, temperature, max_loops, mcp_url, and streaming_on, and a body used as the system prompt. The command builds and lists the agents without running them.
In Swarms 15.0.3, passing a directory to --markdown-path fails with Unsupported file type. Load one file at a time.
swarms heavy-swarm currently fails on every run with a TypeError, because the CLI passes loops_per_agent and random_loops_per_agent to a HeavySwarm constructor that no longer accepts them. Use the Python API instead. swarms llm-council --task "..." works, and needs OpenAI, Google, Anthropic, and xAI keys for its default members.
