Integration

Askli integration

Askli (formerly Chaindesk) builds AI chatbot agents grounded on your own data. Ask an agent a question and get its answer with the source chunks used, create, read, update and delete agents and datastores, add datasources to a datastore (a web page, a whole web site from its sitemap, raw text or an uploaded file), run a similarity search over a datastore, and read the conversations and messages your agents handled. Streaming answers and the deprecated datasource upsert, update and file-upload routes are not exposed. Authentication is an API key from the Askli account settings page sent as Authorization: Bearer.

Connect Askli How integrations work

Ask it in chat

Say what should happen.
The workflow builds itself.

When this happens, do that in Askli.Built from the 16 Askli tools below, on the connection you add once.

What you can do with Askli

Each of these is one operation on one endpoint, available both as a workflow step and as a tool an AI agent can call on its own. Nothing bundles several operations into one action, so what a run did stays readable afterwards.

Most endpoints run on the free plan. A few, mainly the ones that publish on your behalf, need a paid plan.

Agent, 4 operations

create_agentCreate agentPOSTCreates an agent (an AI chatbot) with its model, temperature, prompts and visibility. description is required. Returns the agent with its id.
delete_agentDelete agentDELETEDeletes an agent and returns the deleted agent.
get_agentGet agentGETReturns an agent: name, description, model, temperature, prompts and visibility.
update_agentUpdate agentPATCHUpdates an agent. The API requires description on every update: read the agent with get_agent first and send its description back with the change. Leave other fields out to keep them.

Datasource, 4 operations

create_datasourceCreate datasourcePOSTAdds a datasource to a datastore from a web page or a whole web site (type web_page or web_site, with config.source_url and for web_site config.sitemap), or from raw text (type text, with datasourceText).
delete_datasourceDelete datasourceDELETEDeletes a datasource from its datastore and returns the deleted datasource.
get_datasourceGet datasourceGETReturns a datasource with its type, name, status and configuration.
upload_datasource_fileUpload datasource filePOSTAdds a datasource to a datastore by uploading a file (type file). Returns the datasource with its id and status.

Datastore, 4 operations

create_datastoreCreate datastorePOSTCreates a datastore, the knowledge base that holds datasources for agents. description and type (qdrant, the only value) are required.
delete_datastoreDelete datastoreDELETEDeletes a datastore and returns the deleted datastore.
get_datastoreGet datastoreGETReturns a datastore with its name, description and type.
update_datastoreUpdate datastorePATCHUpdates a datastore. The API requires description and type on every update: read the datastore with get_datastore first and send its description back with the change.

Query, 2 operations

query_agentQuery agentPOSTAsks an agent a question and returns its answer, the conversationId and visitorId (pass them back to continue the same conversation) and the source chunks used. A private agent needs the API key.
query_datastoreQuery datastorePOSTReturns the document fragments of a datastore most similar to a query, each with its text, score, source url and datasource id and name. topK caps the number of results.

Conversation, 1 operation

list_conversationsList conversationsGETLists conversations, 20 per page by default (take), paged with cursor (the last conversation id of the previous page), filtered by agent, channel, status, priority or assignee.

Message, 1 operation

list_conversation_messagesList conversation messagesGETLists the messages of a conversation, paged with cursor (the last message id of the previous page) and take, filtered by sender (from), evaluation (eval) or read state.

Use Askli from Claude Code, Cursor or Codex

Connect Askli in LiveContext, create an API key in Settings, then add the LiveContext MCP server to your client. Its Askli operations become tools the client can call, next to every other integration you connected.

Claude Code

claude mcp add --transport http livecontext https://livecontext.ai/mcp --header "X-API-Key: YOUR_API_KEY"

Codex (~/.codex/config.toml)

[mcp_servers.livecontext]
url = "https://livecontext.ai/mcp"

[mcp_servers.livecontext.http_headers]
"X-API-Key" = "YOUR_API_KEY"

Cursor, Claude Desktop and other JSON configs

{
  "mcpServers": {
    "livecontext": {
      "type": "http",
      "url": "https://livecontext.ai/mcp",
      "headers": {
        "X-API-Key": "YOUR_API_KEY"
      }
    }
  }
}

Set up the MCP server step by step

How to connect Askli

Create your workspace
Sign up free. Nothing to install.
Connect Askli once
Add your Askli connection in Settings. Every tool from this integration reuses it, and secrets are encrypted at rest and visible only to you.
Describe the job in chat
The workflow builds itself in front of you and picks the endpoints it needs. Drag the steps yourself on the canvas whenever you would rather be explicit.