Integration

VectorShift integration

VectorShift is a platform for building AI pipelines, chatbots, knowledge bases and code transformations. Run a pipeline (one run or many in parallel) and pause, resume or stop its runs, chat with a deployed chatbot, create knowledge bases, index files, web pages, Wikipedia, YouTube or arXiv content into them and run semantic queries, and create and run Python transformations; list, fetch and delete each kind of object. Creating pipelines and chatbots, and uploading files to a chatbot conversation, are not exposed: their request bodies are not described by the API reference. The vendor labels this REST API a beta developer preview. Authentication is an API key from Profile, API Keys, sent as Authorization: Bearer.

Connect VectorShift How integrations work

Ask it in chat

Say what should happen.
The workflow builds itself.

When this happens, do that in VectorShift.Built from the 27 VectorShift tools below, on the connection you add once.

What you can do with VectorShift

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.

Knowledge base, 4 operations

create_knowledge_baseCreate knowledge basePOSTCreates an empty knowledge base and returns its id; add content with index_knowledge_base.
delete_knowledge_baseDelete knowledge baseDELETEDeletes a knowledge base and its documents permanently.
get_knowledge_baseGet knowledge baseGETFetches one knowledge base by id, or by name (optionally with username or org_name). Returns status and the knowledge base object.
list_knowledge_basesList knowledge basesGETLists your knowledge bases. Returns object_ids, plus the full objects when verbose is true. Pass an id to query_knowledge_base or index_knowledge_base; to fetch one by name use get_knowledge_base.

Transformation, 4 operations

create_transformationCreate transformationPOSTCreates a transformation: a Python function run in VectorShift's isolated environment, with declared input and output schemas. Returns its id for run_transformation.
delete_transformationDelete transformationDELETEDeletes a transformation permanently.
get_transformationGet transformationGETFetches one transformation by id, or by name (optionally with username or org_name). Returns status and the transformation object.
list_transformationsList transformationsGETLists your transformations (and, with include_shared, those shared with you). Returns object_ids, plus the full objects when verbose is true.

Chatbot, 3 operations

delete_chatbotDelete chatbotDELETEDeletes a chatbot permanently.
get_chatbotGet chatbotGETFetches one chatbot by id, or by name (optionally with username or org_name). Returns status and the chatbot object.
list_chatbotsList chatbotsGETLists your chatbots (and, with include_shared, those shared with you). Returns object_ids, plus the full objects when verbose is true. Pass an id to run_chatbot.

Pipeline, 3 operations

delete_pipelineDelete pipelineDELETEDeletes a pipeline permanently.
get_pipelineGet pipelineGETFetches one pipeline by id, or by name (optionally with username or org_name for someone else's). Returns status and the pipeline object.
list_pipelinesList pipelinesGETLists your pipelines (and, with include_shared, those shared with you). Returns object_ids, plus the full objects when verbose is true. Pass an id to run_pipeline.

Run, 3 operations

run_chatbotRun chatbotPOSTSends a message to a chatbot and returns its reply in output_message, with conversation_id (pass it back to continue the same conversation) and any follow_up_questions.
run_pipelineRun pipelinePOSTRuns a pipeline with the given inputs and waits for it: returns status (success or failed), run_id and outputs, a map of the pipeline's output names to their values.
run_transformationRun transformationPOSTRuns a transformation with the given inputs and returns status and result, the function's outputs.

Document, 2 operations

delete_knowledge_base_documentsDelete knowledge base documentsDELETEDeletes documents from a knowledge base, the ids given as one comma-separated text. Same effect as remove_knowledge_base_documents, which takes a list.
find_knowledge_base_documentsFind knowledge base documentsGETFinds documents in a knowledge base by title, integration or folder and returns their document_ids.

Terminate, 2 operations

terminate_chatbot_runTerminate chatbot runPOSTStops a chatbot's running pipeline run. Returns status.
terminate_pipeline_runTerminate pipeline runPOSTStops a running pipeline run. Returns status.

Bulk run, 1 operation

bulk_run_pipelineBulk run pipelinePOSTRuns the same pipeline several times in parallel, one run per entry of runs. Returns status and run_outputs, one result per run.

Delete, 1 operation

remove_knowledge_base_documentsRemove knowledge base documentsPOSTDeletes documents from a knowledge base, the ids given as a list. Same effect as delete_knowledge_base_documents, which takes comma-separated text.

Index, 1 operation

index_knowledge_baseIndex knowledge basePOSTAdds content to a knowledge base: send exactly one of url_data (a web page, optionally crawled), wikipedia, youtube, arxiv or file_data (a base64 file). Returns the new document_ids.

Pause, 1 operation

pause_pipeline_runPause pipeline runPOSTPauses a running pipeline run; resume it with resume_pipeline_runs. Returns status.

Query, 1 operation

query_knowledge_baseQuery knowledge basePOSTSearches a knowledge base with one or more queries and returns status and result. search_metadata sets top_k and filters; config turns on reranking, query transformation or a generated answer.

Resume, 1 operation

resume_pipeline_runsResume pipeline runsPOSTResumes one or more paused runs of a pipeline. Returns status.

Use VectorShift from Claude Code, Cursor or Codex

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

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