Integration

ChatBotKit integration

ChatBotKit is a platform for building AI agents (bots) with their own backstory, model, knowledge datasets and skillsets of abilities. Create and update bots, start conversations and get the bot's reply (stateful or one-off), read and rate messages, manage datasets and their records and files, search a dataset, manage skillsets and abilities and run an ability, keep contacts and long-term memories, schedule tasks and read their runs, organise resources in blueprints and spaces, ask typed decision questions about a text, and read usage, models and event logs. Conversations, contacts and memories hold end-user personal data. Not exposed: deleting (blueprints aside), cloning, syncing, triggering and cancelling (those routes take an empty JSON body this connector does not send), listing a blueprint's resources (the list includes channel integrations, whose settings hold tokens), background dispatch and live subscriptions (streaming), channel integrations (Slack, WhatsApp, Telegram and others, whose settings hold third-party tokens), secrets, users and their tokens, portals, policies, computers, space storage and sites, exports and the GraphQL endpoint. Authentication is an API token created at chatbotkit.com/tokens, sent as a Bearer token.

Connect ChatBotKit How integrations work

Ask it in chat

Say what should happen.
The workflow builds itself.

When this happens, do that in ChatBotKit.Built from the 97 ChatBotKit tools below, on the connection you add once.

What you can do with ChatBotKit

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.

List, 29 operations

list_blueprintsList blueprintsGETLists blueprints (projects that group bots, datasets, skillsets and other resources).
list_botsList botsGETLists the account's bots (name, model, backstory, connected dataset and skillset), newest first, a page at a time with cursor and take.
list_contact_conversationsList contact conversationsGETLists a contact's conversations. Personal data.
list_contact_memoriesList contact memoriesGETLists the memories stored about a contact. Personal data.
list_contact_ratingsList contact ratingsGETLists the ratings (votes with value and reason) a contact gave to bots, conversations or messages.
list_contactsList contactsGETLists contacts (the end users the bots talk to). Personal data: names, emails, phone numbers.
list_contact_spacesList contact spacesGETLists the spaces that belong to a contact.
list_contact_tasksList contact tasksGETLists the tasks scheduled for a contact.
list_conversation_attachmentsList conversation attachmentsGETLists the files attached to a conversation (name, MIME type, size).
list_conversation_messagesList conversation messagesGETLists the messages of a conversation (type user, bot and others, and text), newest first by default. Personal data.
list_conversationsList conversationsGETLists conversations, newest first, a page at a time with cursor and take. Conversations hold end-user messages: personal data.
list_dataset_filesList dataset filesGETLists the files attached to a dataset as record sources.
list_dataset_recordsList dataset recordsGETLists the records (text chunks) of a dataset, a page at a time with cursor and take.
list_datasetsList datasetsGETLists the account's datasets (knowledge bases used by bots), with their search settings.
list_event_logsList event logsGETLists event log entries (event type and the related conversation, bot, contact, task and other ids), newest first.
list_filesList filesGETLists the account's files (name, description, visibility), a page at a time.
list_magic_promptsList magic promptsGETLists the magic prompts (generators for descriptions, records, abilities and more) available for generate_magic_from_prompt.
list_memoriesList memoriesGETLists the long-term memories bots have stored (facts about contacts), a page at a time. Personal data.
list_platform_abilitiesList platform abilitiesGETLists the ready-made ability templates of the ChatBotKit platform, with their instruction and input schema.
list_platform_actionsList platform actionsGETLists the platform actions abilities can use, with usage examples.
list_platform_examplesList platform examplesGETLists the ready-made bot examples of the ChatBotKit platform.
list_platform_modelsList platform modelsGETLists the AI models available on ChatBotKit (id to use as model, provider, family, token limits).
list_skillset_abilitiesList skillset abilitiesGETLists the abilities of a skillset with their instruction and linked bot, file, space or secret id.
list_skillsetsList skillsetsGETLists the account's skillsets (groups of abilities a bot can use).
list_spacesList spacesGETLists spaces (storage areas where bots keep files).
list_task_executionsList task executionsGETLists the runs of a task with their status, outcome, summary, failure reason and conversation.
list_tasksList tasksGETLists tasks (scheduled instructions a bot carries out) with their schedule, status and next run.
list_teamsList teamsGETLists the teams the account belongs to.
list_usage_recordsList usage recordsGETLists individual usage records (type, count, and the related conversation, bot or task).

Fetch, 18 operations

fetch_blueprintFetch blueprintGETReturns one blueprint with its name, description and visibility.
fetch_botFetch botGETReturns one bot with its backstory, model, dataset, skillset, visibility and features.
fetch_bot_usageFetch bot usageGETReturns a bot's usage totals (tokens, conversations, messages), optionally between two dates.
fetch_contactFetch contactGETReturns one contact with name, email, phone, nickname and preferences. Personal data.
fetch_conversationFetch conversationGETReturns one conversation with its bot, model, backstory, dataset, skillset, contact and expiry.
fetch_conversation_messageFetch conversation messageGETReturns one message of a conversation with its type and text.
fetch_conversation_usageFetch conversation usageGETReturns a conversation's usage totals (tokens and messages), optionally between two dates.
fetch_datasetFetch datasetGETReturns one dataset with its name, description and search settings.
fetch_dataset_recordFetch dataset recordGETReturns one dataset record with its text, source and metadata.
fetch_fileFetch fileGETReturns one file's details (name, description, visibility).
fetch_memoryFetch memoryGETReturns one memory with its text, contact and bot.
fetch_platform_exampleFetch platform exampleGETReturns one platform example with its full details.
fetch_skillsetFetch skillsetGETReturns one skillset with its name, description, visibility and state.
fetch_skillset_abilityFetch skillset abilityGETReturns one ability with its instruction and linked resources.
fetch_spaceFetch spaceGETReturns one space with its name, description and links.
fetch_taskFetch taskGETReturns one task with its schedule, state, last and next run, status and outcome.
fetch_usageFetch usageGETReturns the account's usage in the current billing period: tokens, conversations, messages, and counts of datasets, records, skillsets, abilities and files.
fetch_usage_seriesFetch usage seriesGETReturns daily usage for the last 90 days: tokens, conversations and messages per day.

Create, 14 operations

create_blueprintCreate blueprintPOSTCreates a blueprint (a project grouping resources) and returns its id.
create_botCreate botPOSTCreates a bot (an AI agent) with a name, backstory (its instructions), model and optional dataset and skillset, and returns its id.
create_contactCreate contactPOSTCreates a contact (an end user) with a fingerprint, name, email, phone or nickname, and returns its id. Personal data.
create_conversationCreate conversationPOSTCreates a conversation, usually for a bot (botId) or with its own model and backstory, optionally with first messages, and returns its id.
create_conversation_messageCreate conversation messagePOSTAdds a message of a given type (user, bot, context, instruction...) to a conversation's history without a bot reply.
create_datasetCreate datasetPOSTCreates a dataset (a knowledge base a bot searches when answering) and returns its id. Search settings such as searchMaxRecords are optional.
create_dataset_recordCreate dataset recordPOSTAdds a record (a piece of text with an optional source and metadata) to a dataset and returns its id.
create_decisionCreate decisionPOSTAnswers typed questions about a text or JSON state with a model: each question is boolean, choice or score, keyed by a name, and the answers come back under the same names.
create_fileCreate filePOSTCreates a file entry (name, description, visibility) and returns its id; send its content next with upload_file.
create_memoryCreate memoryPOSTStores a long-term memory (a fact as text), optionally for a contact and a bot, with an optional expiry, and returns its id.
create_skillsetCreate skillsetPOSTCreates a skillset (a group of abilities, the tools a bot can call) and returns its id.
create_skillset_abilityCreate skillset abilityPOSTAdds an ability (an instruction the bot can run as a tool, optionally linked to a bot, file, space or secret) to a skillset and returns its id.
create_spaceCreate spacePOSTCreates a space and returns its id.
create_taskCreate taskPOSTCreates a task: an instruction (name and description) a bot performs on a schedule (cron or date, in a timezone), optionally for a contact, and returns its id.

Update, 13 operations

update_blueprintUpdate blueprintPOSTUpdates a blueprint's name, description or visibility. Only the fields sent are changed; the others keep their values.
update_botUpdate botPOSTUpdates a bot's name, backstory, model, dataset, skillset, privacy, moderation or visibility. Only the fields sent are changed; the others keep their values.
update_contactUpdate contactPOSTUpdates a contact's name, email, phone, nickname, fingerprint or preferences. Only the fields sent are changed; the others keep their values. Personal data.
update_conversationUpdate conversationPOSTUpdates a conversation's name, bot, model, backstory, dataset, skillset, contact or expiry. Only the fields sent are changed; the others keep their values.
update_conversation_messageUpdate conversation messagePOSTChanges the text, type or metadata of a stored message. Only the fields sent are changed; the others keep their values.
update_datasetUpdate datasetPOSTUpdates a dataset's name, description or search settings. Only the fields sent are changed; the others keep their values.
update_dataset_recordUpdate dataset recordPOSTChanges the text, source or metadata of a dataset record and re-indexes it. Send only the fields to change; the others keep their values.
update_fileUpdate filePOSTUpdates a file's name, description, visibility or blueprint. Only the fields sent are changed; the others keep their values.
update_memoryUpdate memoryPOSTUpdates a memory's text, contact, bot or expiry. Only the fields sent are changed; the others keep their values.
update_skillsetUpdate skillsetPOSTUpdates a skillset's name, description, visibility or state; its abilities are not changed. ChatBotKit does not document whether fields left out are kept: send every field you want to keep.
update_skillset_abilityUpdate skillset abilityPOSTUpdates an ability's name, description, instruction, links or blueprint. Only the fields sent are changed; the others keep their values.
update_spaceUpdate spacePOSTUpdates a space's name, description, blueprint or contact. Only the fields sent are changed; the others keep their values.
update_taskUpdate taskPOSTUpdates a task's instruction, schedule, bot, contact, limits or state (enabled or disabled). Only the fields sent are changed; the others keep their values.

Search, 6 operations

search_bot_memorySearch bot memoryPOSTSearches the long-term memories a bot has stored for a keyword or phrase and returns the matching memories.
search_contact_memorySearch contact memoryPOSTSearches the memories stored about a contact for a keyword or phrase. Personal data.
search_datasetSearch datasetPOSTSearches a dataset for records relevant to a query and returns them with their score, optionally filtered on record metadata.
search_memorySearch memoryPOSTSearches all long-term memories for a keyword or phrase, optionally for one contact or bot.
search_platform_abilitiesSearch platform abilitiesPOSTSearches the ready-made ability templates for a keyword or phrase.
search_platform_examplesSearch platform examplesPOSTSearches the ready-made bot examples for a keyword or phrase.

Complete, 2 operations

complete_conversation_messageComplete conversation messagePOSTSends a user message (text) to a stored conversation and returns the bot's reply, which is saved in the conversation history.
complete_stateless_conversationComplete stateless conversationPOSTGets the bot's next reply to a list of messages without storing a conversation: pass the messages (type user or bot, text) and a botId or a model and backstory.

Downvote, 2 operations

downvote_conversationDownvote conversationPOSTRates a conversation negatively (value, -100 by default in ChatBotKit) with an optional reason.
downvote_conversation_messageDownvote conversation messagePOSTRates one message negatively (value, -100 by default in ChatBotKit) with an optional reason.

Upvote, 2 operations

upvote_conversationUpvote conversationPOSTRates a conversation positively (value, 100 by default in ChatBotKit) with an optional reason.
upvote_conversation_messageUpvote conversation messagePOSTRates one message positively (value, 100 by default in ChatBotKit) with an optional reason.

Attach, 1 operation

attach_dataset_fileAttach dataset filePOSTAttaches a file to a dataset as a source (type source), so its content can be turned into records.

Compact, 1 operation

compact_messagesCompact messagesPOSTSummarises a list of messages into one compact text, to keep a long exchange within the model's context.

Delete, 1 operation

delete_blueprintDelete blueprintPOSTDeletes a blueprint. With deleteResources true, every resource in it (bots, datasets, skillsets...) is deleted too; with false only the blueprint goes.

Detach, 1 operation

detach_dataset_fileDetach dataset filePOSTDetaches a file from a dataset; deleteRecords decides whether the records made from it are deleted too.

Download, 1 operation

download_fileDownload fileGETReturns a URL from which the file's content can be downloaded.

Ensure, 1 operation

ensure_contactEnsure contactPOSTReturns the id of the contact with this fingerprint (your own user id), creating it if it does not exist. Personal data.

Execute, 1 operation

execute_skillset_abilityExecute skillset abilityPOSTRuns one ability with an input (text, JSON or YAML) and returns its result, any error, the messages it produced and token usage. The ability may act on linked services.

Generate, 1 operation

generate_magic_from_promptGenerate magic from promptPOSTGenerates text with a magic prompt (for example a bot description or dataset records) from an input text, and returns it with token usage.

Send, 1 operation

send_conversation_messageSend conversation messagePOSTAdds a user message to a conversation without asking the bot to reply; use complete_conversation_message to send and get the reply in one call.

Upload, 1 operation

upload_fileUpload filePOSTUploads the content of a file created with create_file, from a public http(s) URL or a data: URL. Files up to 4.5 MB.

Upsert, 1 operation

upsert_conversation_contactUpsert conversation contactPOSTCreates or updates the contact (end user) of a conversation from a fingerprint, name, email or phone, and returns the contact id. Personal data.

Use ChatBotKit from Claude Code, Cursor or Codex

Connect ChatBotKit in LiveContext, create an API key in Settings, then add the LiveContext MCP server to your client. Its ChatBotKit 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 ChatBotKit

Create your workspace
Sign up free. Nothing to install.
Connect ChatBotKit once
Add your ChatBotKit 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.