Namespace Commands

Namespace commands are the primary names for an agent's actions. Use these names to learn Link's capabilities and write examples whose subject and operation are explicit, such as /task:list.

Declared commands

This list covers every action with the command surface in the agent and model registries consumed by Link Harness. Each entry links to its owning declaration. It includes actions requiring a provider or grant; /help shows the subset available to the current agent. Root-level /help is a discovery command outside these namespaces.

Arguments and availability

Link Harness handles conversation controls and model selection locally, as described in Commands. Actions reached through a route take a JSON object after their name, using that action's request contract. A declaration alone does not guarantee a working provider; the agent's grants, settings and connected providers determine what runs. See Action Namespaces.

Familiar spellings

Command Aliases lists the compatibility spellings that resolve to these commands. An alias uses its target's implementation, arguments and availability rules. Native commands offered by another harness belong to that harness and are outside this inventory.