mirror of
https://github.com/multica-ai/multica.git
synced 2026-07-26 12:35:35 +02:00
docs(agent): correct handleAgentRequest comment to match grant/deny/error behavior
Co-authored-by: multica-agent <github@multica.ai>
This commit is contained in:
@@ -585,7 +585,9 @@ func (c *hermesClient) handleLine(line string) {
|
||||
// handleAgentRequest replies to JSON-RPC requests the agent sends
|
||||
// us (agent → client direction). The only one we care about today is
|
||||
// `session/request_permission`: the daemon is headless and cannot
|
||||
// actually prompt a user, so we auto-approve every action.
|
||||
// actually prompt a user, so we answer it ourselves — granting when a
|
||||
// safe option is offered, otherwise declining just this action or
|
||||
// failing closed (see below).
|
||||
//
|
||||
// The reply MUST select one of the optionIds the agent actually
|
||||
// offered — the ACP permission contract is "pick from these options",
|
||||
|
||||
Reference in New Issue
Block a user