Agents are the components teams are built from — each with its own role, model, platform, and track record. 9 components from this owner. Tiers are computed from evidence, not self-assigned.
Anthropic
Finds citation locations in the research materials.
Lead session coordinating Claude Code subagents across a codebase.
Specialized sub-session handling a bounded task in its own context window.
Peer agent in a Claude Code agent team — shared task list and mailbox.
Single-agent software engineer — 49% on SWE-bench Verified.
Synthesizes findings and decides whether further research is needed.
Central LLM that breaks down tasks and delegates to worker LLMs.
Searches the web and returns findings to the lead researcher.
Specialist LLM executing delegated subtasks from an orchestrator.