Skip to content llms.txt

invocation_patterns

How agents are triggered. Mechanisms include natural language detection, @-mention syntax, slash commands, CLI flags, or automatic delegation.

Type: object Content type: agents

ProviderSupportedMechanism
claude-codenatural language (Claude auto-delegates), @agent-<name> @-mention, or --agent flag at startup for session-wide agent; 3 patterns
codexspawn-agent tool at LLM discretion; LLM selects role from dynamically-built role descriptions combining built-in and user-defined roles
copilot-cli/agent slash command, natural language instruction, auto-inference (suppressible with disable-model-invocation), or --agent flag; 4 patterns
cursorCustom agents are invoked through the Cursor IDE agent picker; background-enabled agents can also run concurrently via the background agents UI.
factory-droidexposed as subagent_type targets for the Task tool; droids enabled by default with live progress reporting in UI
kirokeyboard shortcut binding for IDE invocation; welcomeMessage shown on agent start; manual selection via agent picker
opencodeAgent name derives from the filename (configEntryNameFromPath). Agents with mode: primary can be selected as the session agent; agents with mode: subagent are invoked by other agents via description-driven matching; mode: all allows both.
roo-codeUsers switch modes via the mode selector or /mode command; each mode is a distinct operating persona with its own roleDefinition and tool groups
crushnot documented
windsurfnot documented
zednot documented