MCP server
MCP tools
Tools exposed by the KonvoAI MCP server for reading and acting on conversations.
All tools run with your KonvoAI identity. Organization-scoped reads and writes enforce the same membership rules as the web app (partners cannot use internal tools).
Read tools
| Tool | Description |
|---|---|
list_organizations | Organizations you belong to, with your role. |
list_conversations | Conversations in an organization; optional status (open, closed, resolved). |
get_conversation | Conversation metadata plus the full ticket event timeline for a streamId. |
list_members | Members of an organization (excluding the partner role). |
list_ai_agents | The organization's AI agents with live status, channels, and schedule. |
get_ai_agent | One AI agent's details, including any staged channel/schedule edit. |
list_channels | The organization's channels — pick launch targets for set_agent_channels. |
get_agent_reply | Poll an agent chat session for the latest reply and its status. |
Write tools
| Tool | Description |
|---|---|
respond_to_conversation | Send a text reply (streamId, body; optional clientToken for idempotency). |
set_conversation_status | Set status to open, closed, or resolved. |
assign_conversation | Assign to a user or ai_agent, or pass null to clear. |
snooze_conversation | Snooze until a Unix ms timestamp, or null to unsnooze. |
pin_conversation | Pin or unpin a conversation. |
message_ai_agent | Send a message to an agent's configurator chat; returns ids to poll with get_agent_reply. |
answer_agent_question | Answer (or dismiss) a question the agent asked mid-conversation. |
set_agent_channels | Stage which channels the agent answers on, plus an optional schedule. |
publish_ai_agent | Publish the agent's draft config — knowledge, tools, staged channels — making it live. |
unpublish_ai_agent | Take the agent off the air; channels and drafts stay untouched. |
create_ai_agent | Create a new (unpublished) AI agent. |
rename_ai_agent | Rename an AI agent. |
delete_ai_agent | Permanently delete an AI agent and its configuration. |
Write tools dispatch through the same server mutators as the KonvoAI web app, including permission checks, billing gates, and outbound send queues for replies. AI-agent management additionally requires a role that can manage operational settings.
Example flow
list_conversationswith{ "status": "open" }.get_conversationwith{ "streamId": "…" }.respond_to_conversationwith{ "streamId": "…", "body": "Thanks for reaching out!" }.
Talking to an AI agent
Agent replies are produced asynchronously by the AI worker, so the chat is a send-then-poll flow:
message_ai_agentwith{ "agentId": "…", "message": "…" }→ returnssessionIdandreplyMessageId.get_agent_replywith{ "sessionId": "…" }— repeat untilstatusis no longerin_progress.- If the reply carries
pendingQuestions, respond withanswer_agent_question, then poll again.
Changes the agent makes in chat land on its draft configuration; publish_ai_agent makes them live.
Tool results are JSON text in the MCP content payload.